Self-guided online therapeutic assessment in substance abuse care: A Conflict Analysis implementation and feasibility case study

Abstract

“Michael” is a 60-year-old Black male receiving substance abuse treatment at a Veterans Affairs hospital. As part of Michael’s care, he participated in an implementation and feasibility trial evaluating Conflict Analysis (CA), a self-guided online therapeutic assessment. CA leverages innovative diagnostic and therapeutic resources to improve client engagement and therapeutic achievement. CA employs narrative formats, encouraging patients to frame experiences within personal vocabulary, and provides personalized feedback automatically extracted from patient responses. CA is self-guided, widely accessible, and has few costs. Paper presents Michael’s protocol, associated therapeutic outcome measures, and psychologist’s evaluation of CA benefits. Michael completed brief online CA deployment containing a wellness-based personality inventory, four narrative tasks, and an interactive feedback template. Michael completed insight, stress, depression, wellbeing, and perceived diagnostic and therapeutic benefit measures at baseline, post, and three-week follow-up. Interventions took 2.5 h. Attending psychologist reviewed CA protocols and completed diagnostic and therapeutic questionnaires. Michael’s record identifies elevated dominance. Protocol highlights Michael’s drinking as mechanism to lessen social anxieties, anger, and competitiveness. Results suggest CA encouraged increased motivation to change, treatment engagement, and reduced depression. Psychologist rated benefit at maximum threshold. Implications include improving assessment efficacy, increasing online care opportunities, and expanding CA research.

1. Introduction

Single case study research is a central mechanism of psychotherapeutic exploration and innovation. This intimate format provides a candid vantage to showcase clinical dynamics and illustrate methodological nuance. As such, as psychotherapeutic practice has evolved, this method has similarly been refined to demonstrate new practices and techniques. Following this trend, this paper uses the single case design method to showcase new developments in self-guided online formats, shining light on the computer facilitated clinical encounter.

Adapting the classical classic single case design method, Finn and colleagues deployed this research format to demonstrate the benefits of Therapeutic Assessment (TA; Engelman et al., 2016; Finn, 2003; Smith et al., 2010), a model that premises that assessment can, when coupled with supportive feedback, provide therapeutic benefit in addition to diagnostic benefit. Responding to Finn and colleagues’ precedent, we published (M. Levis & Levis, 2020a) an initial online single case study evaluating usage of Conflict Analysis (CA), a self-guided online therapeutic assessment. In contrast to the previous study, in which the only information we had about the participant was on-line self-report materials, we now present a case from a clinical context where we have access to the patient’s psychiatric history as well as an attending psychologist’s feedback about the intervention. Whereas our prior study laid the groundwork for evaluating online case studies, the present study maps the feasibility and implementability of this method in a clinical context. The patient, “Michael”, is a 60-year-old Black male, who is divorced, a Marine Veteran, has an Associate’s degree, and receiving care in an residential New England Veterans Affairs (VA) substance use recovery center. Michael’s case was chosen at random from a sample of 13 patients that enrolled and provided informed consent for an implementation and feasibility trial investigating CA’s diagnostic and therapeutic benefits. CA is a self-guided online diagnostic and therapeutic tool. CA includes the Relational Modality Evaluation Scale (RMES; A. J. Levis, 1988), an interpersonal screening and intensity inventory, and the Metaphor Tasks, a sequence of interactive drawing and writing prompts. CA automatically generates a diagnostic record and framework for changes based upon multiple-choice selections, creative responses, and self-reflective writing. Output is presented upon segment conclusion and in the Feedback Profile, a cumulative summary presented upon completion. The Feedback Profile organizes content automatically extracted from patients’ responses and encourages reflection on the extracted content, providing space for written reflections. The Feedback Profile, including patients’ follow-up reflections, are available for download and forwarding to consulting clinicians.

CA functions diagnostically by guiding patients to identify how they deal with conflict. The primary approaches to dealing with conflict are conceptualized as differing along the axes of power (dominance versus submissiveness) and attitude (cooperation versus antagonism). CA functions therapeutically by helping patients document their pattern of dealing with conflict and initiate steps to improve upon this pattern. CA integrates techniques associated with narrative disclosure (Pennebaker, 2000) and psychoeducation (Taylor-Rodgers & Batterham, 2014).

In contrast to Therapeutic Assessment, which utilizes conventional assessment procedures such as the Rorschach (Exner, 2003) and the MMPI (Ben-Porath & Tellegen, 2008) alongside supportive and exploratory feedback, CA instead relies on measures that have been specifically developed for both diagnostic and therapeutic benefit (M. Levis et al., 2020; M. Levis & Levis, 2020b), coupled with feedback automatically culled from user responses. In contrast to Therapeutic Assessment, CA is online and self-guided, a format that may reduce the time and cost burden associated with in-person assessment delivery and evaluation (Yates & Taub, 2003).

As CA functions diagnostically and therapeutically, it bears resemblance to both assessment and intervention models. It should be addressed, however, as noted in the TA literature (Finn & Tonsager, 1997), that assessments can in and of themselves offer therapeutic benefit. A central example of this is the Core-Conflictual Relationship Theme (CCRT; (Luborsky & Crits-Christoph, 1998), an unstructured clinician-rated transference focused assessment. In contrast to CCRT, which requires extensive clinical time, to identify and interpret transference themes, and is associated with marked financial costs (Barber et al., 1998), CA offers a self-guided online resource that leverages clients’ creativity and self-reflection. In contrast to Rorschach (Exner, 2003; Meyer et al., 2017) which also addresses creative associations, CA encourages patients to be creative, authoring their own story (Roe & Davidson, 2005), rather than just being interpretative. CA empowers individuals to extract meaning from their creativity, discern how they resolve conflict, and learn how to improve their behavioral patterns (M. Levis & Levis, 2021).

CA’s theoretical and applied method bears resemblance to Multiple Attribute Decision Making (Yoon & Hwang, 2003), an approach that captures decision making preferences through ranking contrasting decision alternatives. Both methods leverage behavior, logic, and physics to conceptualize psychological patterns and utilize repeated pattern detection to strengthen self-awareness (A. J. Levis, 1988b; Zanakisa et al., 1998). The models, however, differ in that, while MADM is theory agnostic, CA relies on interpersonal theory to model alternative patterns. Practically, the models differ in that, while MADM is data driven, leading users to weigh the utilities of decision vectors, CA draws from narrative and creative projection, encouraging self-reflection and self-analysis. Both models tie in with related psychotherapeutic approaches, including psychodynamic and behaviorist models (Fantino, 1998; Shedler, 2010), respectively, offering alternative care trajectories that remain widely used and broadly researched.

This paper presents CA’s utility as a standalone web-based tool without additional therapeutic services, elucidating CA implementation feasibility and patient experience. It avoids engaging in extended case analysis, allowing readers to encounter Michael’s case record. Within this implementation, an adapted short format version of CA was utilized, consisting of an adapted brief version of the Relational Modality Evaluation Scale (B-RMES; see Online Supplement Table 1 for scale items), a selection of four Metaphor Tasks, and the Feedback Profile. To evaluate CA’s impact, the study will present a range of outcome measures associated with successful psychotherapy outcome, as well as Michael’s summary feedback, and his psychologist’s completion of a survey about CA utilization. Study protocol and measures were approved by site IRB. Michael completed informed consent and provided a release for protocols to be used for research purposes, including publication.

1.2. The client

This study took place in a VA residential substance use recovery unit. Programming in this facility is typically five weeks and consists of group and individual interventions and activities such as creative arts and recreational sports. Patients were considered eligible for participation if they were receiving care at the hospital’s residential substance recovery unit, were 18 years or older, spoke English, and were considered as not being at high-suicide risk by the attending psychiatrist. Patients were excluded if they were unable to use a computer, unable to provide informed consent, incarcerated, or involuntarily committed. Patients who met inclusion criteria, were provided information about CA during their first week of care. If they were interested, they completed informed consent, baseline diagnostic and outcome measures, and then completed CA. CA was designed to be completed 2.5 h. This time could be split up based on patients’ schedule as longs as was completed within 1-week of completing consent. Upon finishing CA, patients completed post outcome measures. Patients completed follow-up outcome measures three weeks after CA completion. Patients used VA provided computers with Internet access. Study administrators were in proximity to patients to answer technical or logistical questions and to monitor safety and appropriate computer usage but did not provide therapeutic services.

Michael received diagnoses of Post-Traumatic Stress Disorder and Severe Alcohol Use Disorder. His current treatment focuses on substance use recovery. Michael lives in a medium sized New England City with his current girlfriend and remains close with his two young daughters from his previous marriage.

2.1. Materials and methods

CA is the adapted online version of the Conflict Analysis Battery (CAB; Levis, 1988), a network of therapeutic assessments that have been used in clinical practice for more than 30 years. It includes diagnostic prompts that help identify patients’ behavioral patterns as well as therapeutic exercises that help patients gain insight about how to change these patterns. In contrast to CAB, CA is accessible online, provides automatic feedback, and includes updated items and prompts (M. Levis & Levis, 2020b). CA has been subject to several revisions following patient qualitative feedback to improve therapeutic achievement (M. Levis, 2017). As CA is delivered virtually, it further extends the range of research-based therapeutic services, allowing increased access for individuals that may have difficulty accessing in-person care.¹

B-RMES is a Likert-rated self-report assessment that evaluates personality and interpersonal patterns and symptom intensity. B-RMES output is used to inform patients about personal patterns and structure interpretation of the Metaphor Tasks. B-RMES includes The Dominant Cooperative (DC), Dominant Antagonistic (DA), Submissive Cooperative (SC), and Submissive Antagonistic (SA) subscales. Typically, individuals will score highest on a single relational modality subscale, the scale with the highest score indicating primary relational patterns. These constructs map closely with the constructs measured by the Interpersonal Circumplex model (Wiggins, 1982). B-RMES has strong psychometric qualities, including test-retest reliability and internal consistency (average Cronbach’s alpha = .89) (M. Levis, 2017).

Metaphor Tasks are a sequence of four creativity and narrative-writing prompts. Tasks begin with an open-ended prompt, allowing patients to follow his or her creative impulse, feeling comfortable exploring topics that might otherwise be difficult to address. Each task includes drawing, writing, and guided self-reflection. Metaphor Tasks follow the premise that the degree to which an individual feels stressed corresponds to the degree that an individual responds to offset the initial stress (A. J. Levis, 1977; M. Levis & Levis, 2020a). While the experience of stress is passive, as one is subjected to a given stressor, the responsive action is active, as it requires personal initiative (A. Levis, 1988a). Each passive/active dialectic presents an illustration of the patient’s conflict resolution pattern, charting the unique ways the person engages stress. Within this implementation, the Metaphor Tasks include two passive/active dialectics. Two of the Metaphor Tasks focus on recalling important experiences. The other two focus on creative writing. Memory focused Metaphor Tasks function to generate emotional arousal, encouraging patients to focus on personal feelings and experiences that are not necessarily openly dealt with. Memory activation is a central method to facilitate emotional arousal and recognized as a “key ingredient” in psychotherapeutic change across treatment modalities (Lane et al., 2015). Creativity focused Metaphor Tasks work to stimulate pattern recognition, aiding patients in detecting their behavioral patterns (L. E. Angus & Mcleod, 2004; Freeman, 2015). On average, completion of each task takes 25 min (M. Levis, 2017).

The Feedback Profile is a summary presentation of the patient’s B- RMES profile and key information extracted from the Metaphor Tasks. Extracted content is grouped into Insight, Motivation to Change, and Personal Growth templates. In addition to presenting the extracted information, each template encourages additional reflection on providing space for written reflections on each template and on the overall experience.

Within this utilization, Michael completed CA on his own using a provided personal computer. Study staff with extensive familiarity with CA were available for logistical support. A unit psychologist was also available to address clinical concerns.

As part of this study’s implementation, Michael completed a range of established diagnostic and outcome measure before commencing CA. Measures included: The Mini International Neuropsychiatric Interview (MINI 7.02; Sheehan et al., 1998), a brief structured diagnostic interview based on the 17 most common DSM-5 psychiatric disorders; The Personal Growth Initiative Scale (PGIS; Robitschek, 1999), a 9-item single factor self-report measuring personal growth initiative. PGIS has adequate internal consistency (Cronbach’s alpha = .77) and has been used internationally in Positive Psychology research (Robitschek, 1999); The Beck Cognitive Insight Scale (BCIS; (Beck et al., 2004), a self-report measure that includes the 9-item Self-Reflectiveness Scale (Cronbach’s alpha = .68) and the 6-item Self-Certainty Scale (Cronbach’s alpha = .60); Perceived Stress Scale (PSS-10 (Cohen et al., 1983); a 10-item self-report stress scale; The Patient Health Questionnaire (PHQ-9; (Kroenke & Spitzer, 2002), a 9-item self-report depression rating scale with strong psychometric qualities, including test-retest reliability, convergent and discriminant validity, and internal consistency (Cronbach’s alpha = .89). The study also included an exploratory 13-item measure that evaluates perceived benefits of completing CA, The Perceived Therapeutic and Diagnostic Benefit Scale (PTDB). The study used two PTDB versions, one that was completed by the patient (PTDB-P) and the other that was clinician-rated (PTDB-C). PTDB-C was completed after patient’s CA completion. Before completing the PTDB-C, the clinician was given a short introduction to CA’s methodology and how to evaluate outcome information. The clinician independently evaluated the patient protocols. The clinician completing the PTDB-C was a clinical psychology intern that had an extensive clinical relationship with enrolled patients. Perceived Psychological Wellbeing (PPW) is an exploratory 7-item self-report scale measuring personal evaluation of psychological wellbeing. Items are as follows: I feel like I understand my behavioral patterns; I feel like I understand my interpersonal patterns; I feel like I understand my personal weaknesses; I feel like I understand my personal strengths; I feel like I know how to make changes in my life; I feel motivated to make changes in my life; I would be interested in assistance to make changes in my life. For more information about measures see Online Supplement.

CA functions by encouraging patients to reflect on personal conflict and conflict resolving patterns, as represented in B-RMES profile and illustrated in the Metaphor Tasks. CA guides patients to contextualize experiences and identify changes within personal taxonomies. Patient scores and narrative responses are extracted and used to further insight, changes, and to help anchor subsequent clinical interventions.

Within Michel’s implementation, after completing baseline measures, he started CA. The treatment plan consisted of him completing the B-RMES, followed by four Metaphor Tasks. In the Feedback Profile, Michael was presented with his scores and responses and asked to evaluate personal patterns and program benefits. Afterwards, Michael completed an evaluation of the benefits associated with CA, which was then repeated at three-week follow-up. As this study deployed CA as a self-guided intervention, Michael completed all parts of CA autonomously.

3.1. Results

3.2. Baseline scores

Michael’s baseline scores on the PHQ-9 depression scale show moderate levels of depression. His PGIS scores are low, indicating reduced motivation to pursue personal changes. Full outcome measure information is presented as follows.

3.3. B-RMES scores

Michael’s B-RMES scores indicate a moderately high Dominant Cooperative profile. He marked moderate agreement with items such as “I have always been a leader and enjoy it” and “I am usually the person who initiates and takes charge of things in my relationships”. In contrast, he marked strong disagreement with items “I find it difficult to say what I feel or what I think for fear of hurting someone” and “I give into others, but I resent them”. His score responses present him as being assertive, vocal, kind, insecure, and sometimes angry. His symptom intensity is low.

3.4. Metaphor tasks

For each of the four tasks, excerpts of Michael’s writings are presented as follows. For ease of presentation, Michael’s responses to specific items are grouped together. Michael’s drawings are presented as follows:

Childhood Memory:

In this reflective task, patients illustrate and write about a significant childhood experience.

Memory:

I was crossing the street following someone I looked up to … I got hit by a car just before I reached the curb. I had a broken right leg and hip and fractured left leg … It was all my fault. I was following someone else. As kids, we pick people that we look up to, someone that I thought was cool. He was just someone from down the block … I stayed in the hospital for 6 months. I failed in school because of it. I just missed too many days of school. I used to have As in school. My cousins pitied me. They thought I could never run again. I was in hospital for 6 months then I was on crutches for a long time. Then one leg outgrew the other, so I had to wear a special heel on my shoe until 9th grade. But after that I started playing sports and I got good.

Reflections and changes:

I didn’t have confidence. I looked up to some other guy … I shouldn’t have looked up to him. I shouldn’t have just run across. I learned not to hesitate … I learned that you got to take your shot … But then I learned confidence. One time I got angry at my dad and then he roughed me up. The next day he told me I had to move out, drove me to my grandmother’s house. After that, I joined the army. I joined the army because I wanted to man up. After basic training, I told him what it was like. He said, I didn’t tell you to do it. I realized that I need to be more direct, not just try to do things to please people.

Animal Metaphor:

In this creative task, patients illustrate and write a story about two animals.

Characters:

Deer, 3-years old, male, eager, shy, fast; Dog, 3-years old, male, calm, obedient, trustworthy.

Conversation:

Deer: So we finally meet. Dog: I been on your trail for the last year and a half. I watched you grow from a young buck and now you are going to be a man. Deer: I don’t think you been watching me that long. You must have been watching someone else. Dog: What do you mean? Deer: I practice being stealth. Dog: I also practice stealth and been watching you since I was young. Deer: So what’s this about? Dog: I am interested in how you made it this long. Deer: (instead of taking off and running), I have been going about my business secretly. Dog: Well, why didn’t you run when I approached? Deer: Because I can smell too. I could tell you were not a threat. 2. (Dog looks at him) So you were interested in me? Deer: Yes, I have been curious about you and watching you grow up. Dog: I agree not to follow you. I know that you are just as clever as me.

Reflections and changes:

I identify with the dog. His decision is to try and understand others. I am curious about my surroundings and what goes on around me. If I can learn something about you and you can learn something about me, we can both move on … I also identify with the deer. I want to learn from others, but I don’t mean any harm. The deer wanted to remain unseen and unknown as long as possible. But when he enters into danger he is ready. The dog was trailing the deer, but the deer let him. The deer wasn’t afraid after a certain point … Yes, [this has] definitely [motivated changes]. Be a more objective person. The whole thing about the dog and the deer was about perspective. They both live different lives. One could be out and about. The deer can’t do that. I need to be honest with myself. And not be afraid of rejections. I need to have no hidden agendas. Its like crossing the street, I need to be direct, and being straightforward and cross it. Not stay there stewing. I want to have trust. Definitely. I should be slower to speak and better at listening.

Its like when you meet women. What approach are you going to use. When a man is drinking, and he is under the influence. His courage is built up. He can talk to any women. When he isn’t drinking, he will never speak to the woman he wants. He doesn’t have the courage unless he is drinking. I can talk to a woman when I am not drinking, but we won’t talk on an honest level. When I am drinking, I can talk for real. Have a real conversation. But that is where the problems are. If we are on a level playing field, I could talk honestly. With the dog and the deer, they should talk on that level and maybe become friends. Like with that woman, if you don’t approach her, you might let the woman of your dreams walk away. The animals have opposite patterns, so our approach really matters. Rejection would be the same, so you got to pick your battles, but you are better off approaching.

I have made a lot of bad decisions. Instead of seeing things for what it is or trying to search for what it is, I become stealth and avoid my true emotions. I need to see that rejection is going to be rejection and shouldn’t be afraid. Sometimes I can’t see the forest from the fucking trees. We always want the best and safest result, but I need to learn to be honest and learn to take rejection, not hide. The buck didn’t ever want to be seen, he is in real danger. But he needs to know that he can trust some people, not be too anxious.

Recent Memory:

In this reflective task, patients illustrate and write a story about a significant recent experience.

What happened:

We had lost one game. The other team was undefeated. My roommate was the star player on the undefeated team. I was the second best on my team, down by one game … I was talking trash against the other folks … We were incarcerated at the time. There was hype around the jail. The Thursday night before the Sunday game, my roommate stopped talking to me. That added fuel to the fire. I started talking even more trash. I had 14 points in the first 12 min. We were down by 3 points, and my roommate tried to shoot, and I got the ball, and was about to shoot when then my roommate screamed, “don’t let him take another shot”. And then someone knocked me hard and I went down. Someone tried to reset me leg but it wouldn’t set. My roommate said “get up old man”. But then I went to the infirmary, and they told me it was ruptured. I went in for surgery but the whole time, I was so angry at my roommate and wanted to stab him. He tried to apologize, but I didn’t want to hear it. He was like “I didn’t mean for that to happen”, but he had said it. It took me out of commission for 11 months. I could barely walk.

Reflections and changes:

I didn’t feel any ill will towards the guy that hit me. I just thought it was a foul. The guy I felt angry at was my roommate who had told him to do it. He had given the order. I was angry at him. I was enraged. I wanted to make him feel what I was feeling at any cost. I didn’t forgive him … I got the short end of the big deal. We eventually played again. And eventually became friends again, but he would still make fun of me, call me old man. How long could I stay mad for anyway? Things like that happen and we move on. I don’t take well about being threatened. Basketball is about being focused, and talking trash is part of the game. I have been in a bunch of jails and I have had to fight. It makes you defensive, because you have to protect yourself. I got to stand my ground, but I have learned that there are alternative ways to get out of shit … Once I told a dude, if you say that again, I am going to punch you in the face. And then I did, and I knocked his teeth out and I was put in solitary confinement for a long time. It happened on the same court. Someone else, later, told me, I had seen you being tough before and challenged me. So I punched him out. People always said that I am territorial, but it’s not that. It’s just that I couldn’t handle guys running their mouth. When I talk trash it’s about playing ball, not about who the person is. I want to work on how not to go from 0 to 100. I want to go from 0 to 15 instead.

Short Story Metaphor:

In this creative task, patients write and illustrate a short story.

Characters:

Me, 59-years old, male, nervous about being on the water; my boss, 38-years old, female, good boss, experienced, runs the show.

What happened:

Going sailing with the crew from the restaurant where I worked at the time. It was five of us, me and my boss. The morning was calm and I was a little nervous being that I cannot swim and being in a sailboat for the first time ever. It turned out to be okay after I got over my fear of the water. I learned how to sail and guide the boat. We went out as far as we could before turning around and heading back to shore. I had been in a boating accident when I was younger, so I don’t fuck around on the water. I learned the dangers of how to stay out of the way of the mast as not to be knocked into the water and the proper technique of docking the sailboat.

Reflections:

I used to be like I don’t like when people get too close to me, or walk up next to me, I would get up in their face. I don’t let anybody invade my space. But one time, one of my facilitators, confronted me, and showed me that I don’t need to retaliate, I can remove myself from the situation, showed me there are other ways of responding to a threat. Showed me that I was responding out of anger. He showed me that I was fooling myself. He showed me that I can create walls, but every wall has a gate to let people in and out. And if you are keeping everything out, nothing is getting in. You can’t get anywhere or lose anything. If you don’t give, you can’t receive.

Changes:

Understanding that I shouldn’t be afraid of the water. Be more willing to take the back seat. Evaluate the situation more. Realize that there are multiple paths. Multiple alternatives. Yes, I am afraid of trying new things. Like, being in rehab. I have never been in rehab before. It’s like coming in from the rain. It’s not easy to get out of what I know and try something new. I am still working on that. I can go from 0 to 100 quick, but I am working on ways to find a medium. I want to deal with it not when it gets to 50, but when it gets to 25. I don’t want it to get to 20. I don’t want to feel that I need to resort to violence. Violence is the last resort of a weak man (Fig. 1).

Fig. 1.

The above table presents Michael’s four drawings.

3.5. Feedback profile

In reviewing his metaphor tasks, and the connection between them:

They are all about me. The first is about being more determined. The second is about love. The third is about anger. The fourth is about getting over fear. I can be a caring person, but I need to learn to be patient.

In reviewing whether CA has helped him gain insight overall:

Yes. That I need to be more patient in my thought process. In the first story, I wasn’t patient. In the animal story, I learned to be patient, to love and trust. In the basketball game, I had to learn to forgive, and in the boat, I learned to spread my wings, and not be so interested in myself.

In reviewing the changes that he now wants to make:

Yes. To be more aware of my emotions. I can go slower instead of running and running out of gas. The rabbit and the turtle.

In reviewing the experience overall:

It was eye opening.

Perceived benefit:

Upon concluding CA and three weeks after concluding CA, Michael answered a range of items that evaluate the intervention’s diagnostic and therapeutic benefit (Online Supplement, Table 2, first and second columns). At both intervals, Michael marked “strongly agree” for all items.

Clinician assessed benefits:

After Michael completed CA, his protocols were shared with one of his psychologists (Online Supplement, Table 2, third column). This clinician marked “strongly agree” for all items.

4.1. Discussion

Michael’s protocol suggests that CA may be feasibly implemented within a clinical context. CA’s self-guided online format appears to facilitate a process of both cathartic self-expression and self-reflection. Even without a clinician being personally present, Michael was able to identify major personal patterns and discern active steps to improve upon these patterns. While we are careful not to suggest that CA provides resources equivalent to traditional psychotherapy, our initial results suggest it may amplify the utility of traditional assessment practices.

Michael’s protocol highlights his Dominant Cooperative pattern, documenting his assertive, competitive, and collaborative stance. We simultaneously see him as someone able to “talk trash” on the courts, to punch people in the face, but also as someone that is afraid, both of water and of directly approaching women. While his memories emphasize self-reliance and the danger of trusting other people, his creative tasks showcase his desire to draw nearer to other people rather than build boundaries. He recognizes that his tendency to be “stealth and avoid my true emotions”, although protective, holds him back. CA’s iterative process helps Michael recognize his underlying emotions and appreciate his avoidant patterns. Michael’s anger and drinking suppress his anxieties about engaging with other people, allowing him to more confidently encounter them on his terms.

Michael’s memory tasks detail passive experiences, occurrences that happened to him, as opposed to situations where he took assertive action. In contrast, within the creative episodes, Michael agenticly responds to his circumstances, and shifts towards resolution. This dichotomy supports the hypothesized passive/active dialectic. In contrasting these related sequences, we may note consistent patterns, but also detect how Michael adapts over the course of CA. Intriguingly, in both memory episodes Michael’s legs were hurt, leaving him unable to run away and thus exposed to danger. In the creative tasks, this reflex to run is similarly obstructed, and yet this disability leads to transformation. In the Animal Metaphor, rather than running away from or chasing after the deer, the dog, the character Michael identifies with, states, “I agree not to follow you. I know that you are just as clever as me.” In reflecting on this story, Michael explains that he “should be slower to speak and better at listening.” Similarly, in his Short Story, Michael explained that learning that he “shouldn’t be afraid of the water. Be more willing to take the back seat. Evaluate the situation more.” Rather than trying to run away or fight back, Michael instead changes his approach, recognizing that “violence is the last resort of a weak man.” Although both the Animal Metaphor and Short Story present resolutions, a subtle shift is apparent between the tasks. While in the earlier task, the characters are defined by their “opposite patterns”, in the later task there is a recognition that “every wall has a gate to let people in and out. And if you are keeping everything out, nothing is getting in … If you don’t give you can’t receive.” Michael slowly realizes that “there are other ways of responding to a threat” and that his intensity presents a suboptimal method of dealing with his anxieties.

Intriguingly Michael’s takeaways from his Childhood Memory and Short Story stand in stark contrast, mapping Michael’s overall course of change from beginning CA to its conclusion. Whereas in the first task, Michael gleans that he must become self-reliant and “more direct, not just try to do things to please people”, in the later task he “realize[s] that there are multiple paths. Multiple alternatives.” By seeing the other not as a threat but as an equal, someone that might be equivalently “clever”, or maybe even more capable than himself, as personified by his boss in the final task, we witness a moderation in Michael’s approach, and a shift towards less dominant behaviors.

Michael’s only references alcohol once during CA, explaining “When I am drinking, I can talk for real. Have a real conversation.” Alcohol seemingly allows Michael’s anxiety to be lessened, providing him license to interact more honestly. Via CA, Michael discerns that there are other methods aside from alcohol that allow him to feel this degree of comfort. Indeed, through CA we can appreciate that Michael’s anger and mistrust is a symptom of his broader dominant relational stance; Michael’s need to be autonomous and a leader sets him up to be suspicious of others and quickly confrontational. As a way to support his sobriety, CA recommends addressing this broader relational pattern, encouraging Michael to, following his own guidance, “realize that there are multiple paths”, and lessen his own intensity. Indeed, drinking appears to be a mechanism to moderate his dominant impulses, allowing him composure, comfort, confidence, and trust in himself and others.

In reviewing this case, we note that Michael found CA therapeutic, motivating him to identify his patterns and start making changes (“To be more aware of my emotions. I can go slower instead of running and running out of gas”) and diagnostic (“It was eye opening”). Michael’s psychologist could accordingly focus subsequent interventions on addressing his anxieties, anger, and relational intensity, and helping him make the identified changes.

Michael’s record emphasizes patterns of dominance, a finding that is present in his RMES scores, Metaphor Tasks, and concluding comments. In some capacity, Michael’s desire to allow himself to be more vulnerable, to go slower, to trust more, hint that he is attempting to moderate his dominant tendencies. That being said, while CA may help articulate this dynamic process, it remains challenging to parse boundaries between Michael’s personal dominant tendencies and dominant tendencies that may result from social pressures, forces that may be particularly present within the Veteran community (Lorber & Garcia, 2010). Indeed given Michael’s family dynamics, as noted in his Childhood Memory, and extensive history within military and carceral contexts (Hinojosa, 2010; Michalski, 2017), we can imagine that there may be social pressures to demonstrate dominant behaviors. Other therapeutic mechanisms, including transference-focused interventions like CCRT (Luborsky & Crits-Christoph, 1998); and Emotion-Focused group therapy (L. E. Angus & Mcleod, 2004) may help illuminate this area and provide supportive interpersonal resources.

4.2. Contrasting approaches

Although Michael did not complete other related interventions, it is nonetheless useful to imagine how his experience would have been different with usage of TA or CCRT, for example. Beyond the practical difference of these other interventions being neither self-guided nor accessible online, these alternative interventions have impactful different conceptual and methodological orientations from CA. Although both TA and CCRT rely on patient/therapist relationships more directly than CA, it should be noted that all three interventions advance relational approaches. While CA encourages self-understanding within the context of interpersonal dynamics (A. J. Levis, 1988a), TA centers on utilizing therapeutic alliance to facilitate change (Engelman et al., 2016), and CCRT relies on transference and countertransference dynamics to achieve insight and improved health outcomes (Luborsky & Crits-Christoph, 1998).

If TA had been implemented, we could imagine a context that Michael would have been given an opportunity to reflect on himself vis- a-vis his MMPI scoring profile. This profile would be filtered through his ` therapist’s careful feedback, allowing an opportunity for Michael to evaluate if one of the assessments’ scoring constructs has personal relevance and an opportunity to develop alliance between Michael and his therapist. This could be informative – allowing insight and self-awareness based on recognized diagnostic schemas – or triggering – as associations with diagnostic pathologies can be off-putting or stigmatizing (Witztum et al., 1995). Depending on the therapist’s dexterity, these issues could be leveraged over time to achieve integration and construct a framework for clinical change. This stands in contrast to CA in that CA avoids illness associated diagnostic labels and encourages the patient to define his or her own parameters of relevance, allowing the patient to relay what is important on their own terms. CA similarly lessens reliance on the importance of the therapist’s expertise, in that the patient has greater personal agency for self-reflection. While this has practical impact in terms of encouraging autonomy and personal growth, and offers expediency, and cost-savings, there is potential that the patient’s interpretation could miss ques and reinforce misappraisals.

If CCRT had been implemented, we could imagine a clinical scenario slightly closer to that of CA. As in CA, the patient’s personal voice would have centrality over diagnostic labels (Luborsky & Crits-Christoph, 1998). As in CA, core patterns of conflict and conflict resolution would be conceptualized as a dialectic process. In contrast to CA, however, this process would be specifically facilitated by the therapist; Michael would be encouraged to relay personal experiences that would then be reconstructed and interpreted via his therapist. Whereas TA rests on the validity and utility of the utilized diagnostic assessment, CCRT relies on the therapist’s interpretative dexterity. The therapist would be responsible to decipher the patient’s 1) wishes (W), 2) responses of others to patient’s wishes (RO), and, in turn, 3) responses of patient in relationship to others’ responses (RS), all from the patient’s unstructured recollections (Luborsky & Crits-Christoph, 1998). CCRT utilization manuals anticipate a vast array of interpretative variants (W = 35, RO = 30, RS = 31; Barber et al., 1995), a relative complexity that stands in contrast in the relative simplicity of CA’s model. For example, Michael’s therapist could, after an extensive number of sessions (Barber et al., 1998), have explained that Michael’s W is “to not be hurt”, his RO might be “to be rejecting”, and his RS could be “being out of control”. While this potential interpretation could offer clinical value, it might not account for the breadth of Michael’s conflictual patterns nor be as conceptually simple as, for example, centering Michael’s treatment on his dominant patterns. To this end, we could expect that Michael would complete his CCRT evaluation with a nuanced appreciation of his behavioral patterns, albeit an appreciation that might be less personally understandable than his CA protocol.

4.3. Conclusions and limitations

In addition to being limited by its design as a single case study, other limitations include the brief nature of CA, the impact of coinciding other treatments, and self-report biases. As this case was selected from an implementation and feasibility trial, rather than an effectiveness study, the study prioritized demonstrating CA’s feasibility rather than confirming its effectiveness. Michael appeared to enjoy expressing himself creatively and engaging in self-reflection, a finding that accords with prior CA research (M. Levis & Levis, 2021). Given the limited nature of this study, it is not possible to speculate how Michael would have responded to other intervention formats or to directly compare CA with other assessments or interventions. Given that Michael did not complete a comparison control condition, we cannot evaluate whether he would have had an improved experience in another therapeutic context. That being said, we can assume that conceptually related interventions, including TA (Finn, 2003), CCRT (Luborsky & Crits-Christoph, 1998), Decision Aid guides (Frysak et al., 2017), and Written Exposure Therapy (Sloan et al., 2015), as well as technically related self-guided online interventions (de Graaf et al., 2009; Holter et al., 2020; McDonald et al., 2020), may have offered Michael parallel benefits.

Michael’s case does not provide evidence for CA’s benefit for specific patients or groups of patients. Additional research is necessary to evaluate CA’s utility with patients with different diagnostic criteria and with diverse patient populations. More broadly, it remains challenging to evaluate CA’s efficacy without an active control condition. Findings should not be extrapolated beyond the specific context of this case. While Michael’s case record suggests an experience of increased self- reflection and self-understanding, changes that appear to confirm with outcome measure and clinician ratings, we have little ability to confirm these changes.

Although computers and internet service were provided by the facility within this study, patients’ lack of computer access and/or lack of dexterity with online interfaces could limit CA’s broader utilization. It can be expected that many patients will, unlike Michael, opt against completing CA or fail to complete follow-up questionnaires. Within typical implementations, CA user data is maintained for two-week periods during which time users can return to complete the intervention. Non-completed data is available to users and clinicians, although the Feedback Profile is only completed generated upon completion. Future versions of CA will include natural language processing resources such as topic modeling (Blei et al., 2003) and semantic analysis (Crossley et al., 2017) to offer additional analytical resources.

Although this study did not specifically address the Coronavirus pandemic, it stands to reason that CA may offer heightened utility during this time when clinical assessments and interventions are increasingly offered online. Future studies should address if CA and related online interventions offer resources to promote engagement in mental health and whether cultural/racial background impacts utilization and intervention experience. Study supports continued exploration of CA and related self-guided online psychotherapeutic resources. A randomized controlled trial with sufficient participant size will be conducted to better evaluate CA effectiveness in comparison with other self- guided interventions.

Table 1.

Patient’s outcome results at baseline, intervention completion (post), and three-week follow-up.

	Score		Score Differences		% change
Outcome Measures	Mean	Sum	Baseline to Post	Baseline to Follow-up	Baseline to Post		Baseline to Follow-up
PGIS
Baseline	2.78	25	+26	+25	104%		100%
Post	5.67	51
Follow-up	5.55	50
PHQ9a
Baseline	2.22	20	− 11	− 8	55%		40%
Post	1	9
Follow-up	1.33	12
PSS-10a
Baseline	2.70	27	+2	− 1	7.41%		3.7%
Post	2.90	29
Follow-up	2.60	26
Self-Reflection (BCIS) Baseline	1.88	17	+4	+2	23.53%	11.76%
Post	2.33	21
Follow-up	2.11	19
Self-Certaintya (BCIS)
Baseline	1.83	11	+4	+4	36.36%	36.36%
Post	3.5	15
Follow-up	2.5	15
PPW
Baseline	3.43	24	+4	+2	16.67%	8.33%
Post	4	28
Follow-up	3.71	26

PGIS = Positive Growth Initiative Scale; PHQ9 = Patient Health Questionnaire-9; PSS-10 = Perceived Stress Scale; Self-Reflection and Self-Certainty BCIS = Beck Cognitive Insight Scales; PPW = Perceived Psychological Wellbeing.

^aHigher scores indicate worse symptoms.

Abbreviations:

CA

Conflict Analysis

RMES

Relational Modality Evaluate Scale