Dataset Information

Speech-based markers for posttraumatic stress disorder in US veterans.

ABSTRACT: BACKGROUND:The diagnosis of posttraumatic stress disorder (PTSD) is usually based on clinical interviews or self-report measures. Both approaches are subject to under- and over-reporting of symptoms. An objective test is lacking. We have developed a classifier of PTSD based on objective speech-marker features that discriminate PTSD cases from controls. METHODS:Speech samples were obtained from warzone-exposed veterans, 52 cases with PTSD and 77 controls, assessed with the Clinician-Administered PTSD Scale. Individuals with major depressive disorder (MDD) were excluded. Audio recordings of clinical interviews were used to obtain 40,526 speech features which were input to a random forest (RF) algorithm. RESULTS:The selected RF used 18 speech features and the receiver operating characteristic curve had an area under the curve (AUC) of 0.954. At a probability of PTSD cut point of 0.423, Youden's index was 0.787, and overall correct classification rate was 89.1%. The probability of PTSD was higher for markers that indicated slower, more monotonous speech, less change in tonality, and less activation. Depression symptoms, alcohol use disorder, and TBI did not meet statistical tests to be considered confounders. CONCLUSIONS:This study demonstrates that a speech-based algorithm can objectively differentiate PTSD cases from controls. The RF classifier had a high AUC. Further validation in an independent sample and appraisal of the classifier to identify those with MDD only compared with those with PTSD comorbid with MDD is required.

SUBMITTER: Marmar CR

PROVIDER: S-EPMC6602854 | biostudies-literature | 2019 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Speech-based markers for posttraumatic stress disorder in US veterans.

Marmar Charles R CR Brown Adam D AD Qian Meng M Laska Eugene E Siegel Carole C Li Meng M Abu-Amara Duna D Tsiartas Andreas A Richey Colleen C Smith Jennifer J Knoth Bruce B Vergyri Dimitra D

Depression and anxiety 20190422 7

<h4>Background</h4>The diagnosis of posttraumatic stress disorder (PTSD) is usually based on clinical interviews or self-report measures. Both approaches are subject to under- and over-reporting of symptoms. An objective test is lacking. We have developed a classifier of PTSD based on objective speech-marker features that discriminate PTSD cases from controls.<h4>Methods</h4>Speech samples were obtained from warzone-exposed veterans, 52 cases with PTSD and 77 controls, assessed with the Clinicia ...[more]

PMID: 31006959

Similar Datasets

Project description:ImportancePosttraumatic stress disorder (PTSD) symptom reduction is linked with lower risk of incident type 2 diabetes (T2D), but little is known about the association between PTSD and comorbid T2D outcomes. Whether PTSD is a modifiable risk factor for adverse T2D outcomes is unknown.ObjectiveTo determine whether patients with PTSD who improved and no longer met diagnostic criteria for PTSD had a lower risk of adverse T2D outcomes compared with patients with persistent PTSD.Design, setting, and participantsThis retrospective cohort study used deidentified data from US Veterans Health Administration (VHA) historical medical records (from October 1, 2011, to September 30, 2022) to create a cohort of patients aged 18 to 80 years with comorbid PTSD and T2D. Data analysis was performed from March 1 to June 1, 2024.ExposuresDiagnoses of PTSD and T2D.Main outcomes and measuresThe main outcomes were insulin initiation, poor glycemic control, any microvascular complication, and all-cause mortality. Improvement of PTSD was defined as no longer meeting PTSD diagnostic criteria, per a PTSD Checklist score of less than 33. Entropy balancing controlled for confounding. Survival and competing risk models estimated the association between meeting PTSD criteria and T2D outcomes. Subgroup analyses examined variation by age, sex, race, PTSD severity, and comorbid depression status.ResultsThe study cohort included 10 002 veterans. More than half of patients (65.3%) were aged older than 50 years and most (87.2%) were men. Patients identified as Black (31.6%), White (62.7%), or other race (5.7%). Before controlling for confounding with entropy balancing, patients who no longer met PTSD diagnostic criteria had similar incidence rates for starting insulin (22.4 vs 24.4 per 1000 person-years), poor glycemic control (137.1 vs 133.7 per 1000 person-years), any microvascular complication (108.4 vs 104.8 per 1000 person-years), and all-cause mortality (11.2 vs 11.0 per 1000 person-years) compared with patients with persistent PTSD. After controlling for confounding, no longer meeting PTSD criteria was associated with a lower risk of microvascular complications (hazard ratio [HR], 0.92 [95% CI, 0.85-0.99]). Among veterans aged 18 to 49 years, no longer meeting PTSD criteria was associated with a lower risk of insulin initiation (HR, 0.69 [95% CI, 0.53-0.88]) and all-cause mortality (HR, 0.39 [95% CI, 0.19-0.83]). Among patients without depression, no longer meeting PTSD criteria was associated with a lower risk of insulin initiation (HR, 0.73 [95% CI, 0.55-0.97]).Conclusions and relevanceThe findings of this cohort study of patients with comorbid PTSD and T2D suggest that PTSD is a modifiable risk factor associated with a modest reduction in microvascular complications. Further research is needed to determine whether findings are similar in non-VHA health care settings.

Project description:BackgroundCurrent research in behavioral cardiology reveals a significant association between posttraumatic stress disorder (PTSD) and increased risk for cardiovascular disease and mortality; however, the underlying mechanisms remain poorly understood. We hypothesized that patients with PTSD would exhibit endothelial dysfunction, a potential mechanism involved in the development and progression of cardiovascular disease.Methods and resultsA total of 214 outpatients treated at the San Francisco Veterans Affairs Medical Center underwent tests of endothelial function and evaluation for PTSD. Flow-mediated vasodilation of the brachial artery was performed to assess endothelial function, and current PTSD status was defined by the PTSD Checklist, based on the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition), with a score ≥40. Multivariable linear regression models were used to estimate the association between PTSD status and endothelial function. Patients with PTSD (n=67) were more likely to be male (99% versus 91%, P=0.04) and to have depression (58% versus 8%, P<0.0001) and were less likely to be on an angiotensin-converting enzyme inhibitor (17% versus 36%, P=0.007) or β-blocker treatment (25% versus 41%, P=0.03). Univariate analysis demonstrated that patients with PTSD had significantly lower flow-mediated vasodilation (5.8±3.4% versus 7.5±3.7%; P=0.003); furthermore, lower flow-mediated vasodilation was associated with increasing age (P=0.008), decreasing estimated glomerular filtration rate (P=0.003), hypertension (P=0.002), aspirin (P=0.03), and β-blocker treatments (P=0.01). In multivariable analysis, PTSD remained independently associated with lower flow-mediated vasodilation (P=0.0005).ConclusionsAfter adjusting for demographic, comorbidity, and treatment characteristics, PTSD remained associated with worse endothelial function in an outpatient population. Whether poor endothelial function contributes to the higher risk of cardiovascular disease in patients with PTSD deserves further study.

Dataset Information

Speech-based markers for posttraumatic stress disorder in US veterans.

Publications

Speech-based markers for posttraumatic stress disorder in US veterans.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets