Development and Validation of Computerized Adaptive Assessment Tools for the Measurement of Posttraumatic Stress Disorder Among US Military Veterans

Brenner, Lisa A.; Betthauser, Lisa M.; Penzenik, Molly; Germain, Anne; Li, Jinjun; Chattopadhyay, Ishanu; Frank, Ellen; Kupfer, David J.; Gibbons, Robert D.

doi:10.1001/jamanetworkopen.2021.15707

Cited by 7 publications

(9 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning has the ability to extract more out of data, and it has been used for various medical diagnosis, such as tree-based models in PTSD assessment [ 18 ], naïve Bayes, random forest, and support vector machines in lung cancer prognosis [ 19 ], XGBoost for kidney disease diagnosis [ 20 ]. We developed a comprehensive machine learning framework for predicting future suicidality occurrence, severity, and imminence (see Tables 1 , 2 ).…”

Section: Resultsmentioning

confidence: 99%

Polyphenic risk score shows robust predictive ability for long-term future suicidality

et al. 2022

View full text Add to dashboard Cite

Suicides are preventable tragedies, if risk factors are tracked and mitigated. We had previously developed a new quantitative suicidality risk assessment instrument (Convergent Functional Information for Suicidality, CFI-S), which is in essence a simple polyphenic risk score, and deployed it in a busy urban hospital Emergency Department, in a naturalistic cohort of consecutive patients. We report a four years follow-up of that population (n = 482). Overall, the single administration of the CFI-S was significantly predictive of suicidality over the ensuing 4 years (occurrence- ROC AUC 80%, severity- Pearson correlation 0.44, imminence-Cox regression Hazard Ratio 1.33). The best predictive single phenes (phenotypic items) were feeling useless (not needed), a past history of suicidality, and social isolation. We next used machine learning approaches to enhance the predictive ability of CFI-S. We divided the population into a discovery cohort (n = 255) and testing cohort (n = 227), and developed a deep neural network algorithm that showed increased accuracy for predicting risk of future suicidality (increasing the ROC AUC from 80 to 90%), as well as a similarity network classifier for visualizing patient’s risk. We propose that the widespread use of CFI-S for screening purposes, with or without machine learning enhancements, can boost suicidality prevention efforts. This study also identified as top risk factors for suicidality addressable social determinants.

show abstract

Section: Resultsmentioning

confidence: 99%

Polyphenic risk score shows robust predictive ability for long-term future suicidality

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The CAT-MH has been utilized in outpatient community programs [57], inpatient hospital settings (i.e., emergency rooms) [58], family physician centres that offer mental health services, correctional centres [59], and most notably in primary care settings [60]. The CAT-MH demonstrates high reliability, with test-retest reliability averaging .92 [55,56,61] and high specificity in assessing for potential diagnoses, with Areas Under the Curve (AUC) ranging from .76 for substance use [62] to .94 for depression and PTSD [63]. Measures of specificity for the CAT-MH are similar to other 'gold standard' measures of mental health symptomatology (e.g., Beck Depression Inventory; [55]).…”

Section: Methodsmentioning

confidence: 99%

Walk-in mental health: Bridging barriers in a pandemic

Wellspring,

Ganesh,

Kreklewetz

2024

PLoS ONE

View full text Add to dashboard Cite

‘Single Session Therapy’ (SST) is a service delivery model that seeks to provide an evidence-based, solution-focused, brief intervention within a single therapy session. The stand-alone session affords the opportunity to provide brief psychological interventions while clients await access to longer-term services. The COVID-19 pandemic has adversely impacted individuals’ mental health. However, the majority of research has investigated patient mental health within hospital settings and community organizations that offer long-term services, whereas minimal research has focused on mental health concerns during COVID-19 within an SST model. The primary aim of the study was to measure client experiences of a brief mental health service. The nature of client mental health concerns who access such services at various points during a pandemic was also investigated. The current study utilized client feedback forms and the Computerized Adaptive Testing—Mental Health (CAT-MH) to measure client experiences and mental health concerns. Qualitative analysis of client feedback forms revealed themes of emotional (e.g., safe space) and informational support (e.g., referrals). Clients also reported reduced barriers to accessing services (e.g., no appointment necessary, no cost), as well as limitations (e.g., not enough sessions) of the Walk-in clinic. Profile analysis of the CAT-MH data indicated that clients had higher rates of depression before COVID-19 (M = 64.2, SD = 13.07) as compared to during the pandemic (M = 59.78, SD = 16.87). In contrast, higher rates of positive suicidality flags were reported during the pandemic (n = 54) as compared to before (n = 29). The lower reported rates of depression but higher rate of suicidality during the pandemic was an unanticipated finding that contradicted prior research, to which possible explanations are explored. Taken together, the results demonstrate the positive experiences of clients who access a single session therapy.

show abstract

“…Our first dataset (referred to as the VA dataset) comprises ♥ ❂ ✸✵✹ participants recruited at a Veterans Health Administration facility for an earlier study 40 . Veterans between the ages of 18 and 89 years were recruited with written informed consent.…”

Section: Participants and Data Sourcesmentioning

confidence: 99%

“…Veterans between the ages of 18 and 89 years were recruited with written informed consent. Once eligibility was determined by the study team, participants completed a PTSD-symptom questionnaire from the CAT-MH PTSD item bank 40,41 , comprising ✷✶✶ items, including some items from the PTSD Checklist (PCL-5). Participants were also assessed by licensed doctorate-level mental health providers using the Clinician-Administered Scale for PTSD for Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (CAPS-5), which resulted in ✽✻ PTSD diagnoses, and ✷✶✽ participants deemed as not having PTSD.…”

Section: Participants and Data Sourcesmentioning

confidence: 99%

Generative AI To Recognize Response Fabrication in PTSD

Chattopadhyay,

Lee,

Evans

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Fabricating symptoms of Post-Traumatic Stress Disorder (PTSD) can hinder accurate clinical assessments via structured diagnostic interviews1,2. Symptom simulation or fabrication is a known problem3,4 in PTSD assessments, with diverse motivations including unmet mental health issues, varied socio-economic factors and the potential for external gain from positive diagnoses. Here we introduce an artificial intelligence (AI) framework referred to as Algorithm VeRITAS (Vetting Response Integrity from cross-Talk in Adversarial Surveys), for detecting symptom fabrication in the context of PTSD diagnosis. In contrast to current approaches to fabrication detection which indirectly assess atypical symptom presentations, and have limited reliability, VeRITAS infers statistical dependencies inherent in true response patterns, flagging responses which violate these subtle constraints. With a study sample of n = 651 patients, VeRITAS has an Area Under the Curve (AUC) of ≧ 0:95 ± 0:02, with sensitivity > 95%, specificity > 88%, and positive likelihood ratio between 9:9 - 19:77. Additionally, VeRITAS is difficult-to-impossible to beat with coaching or training; we demonstrate that having advanced training in mental health diagnosis is not helpful in defeating the algorithm. Our tool offers an objective, diseasespecific, fast (average time ≦ 4 min) detection of simulated or feigned PTSD, and on wider adoption, can potentially help resources and disability concessions reach those genuinely in need, while helping to maintain integrity of clinical data. Moreover, reliably identifying patients who might be fabricating symptoms due to unmet mental health needs or socio-economic compulsions can ultimately improve outcomes in disadvantaged communities.

show abstract

Development and Validation of Computerized Adaptive Assessment Tools for the Measurement of Posttraumatic Stress Disorder Among US Military Veterans

Cited by 7 publications

References 43 publications

Polyphenic risk score shows robust predictive ability for long-term future suicidality

Polyphenic risk score shows robust predictive ability for long-term future suicidality

Walk-in mental health: Bridging barriers in a pandemic

Generative AI To Recognize Response Fabrication in PTSD

Contact Info

Product

Resources

About