Enhancing post-traumatic stress disorder patient assessment: leveraging natural language processing for research of domain criteria identification using electronic medical records.

BACKGROUND: Extracting research of domain criteria (RDoC) from high-risk populations like those with post-traumatic stress disorder (PTSD) is crucial for positive mental health improvements and policy enhancements. The intricacies of collecting, integrating, and effectively leveraging clinical notes for this purpose introduce complexities. METHODS: In our study, we created a natural language processing (NLP) workflow to analyze electronic medical record (EMR) data and identify and extract research of domain criteria using a pre-trained transformer-based natural language model, all-mpnet-base-v2. We subsequently built dictionaries from 100,000 clinical notes and analyzed 5.67 million clinical notes from 38,807 PTSD patients from the University of Pittsburgh Medical Center. Subsequently, we showcased the significance of our approach by extracting and visualizing RDoC information in two use cases: (i) across multiple patient populations and (ii) throughout various disease trajectories. RESULTS: The sentence transformer model demonstrated high F1 macro scores across all RDoC domains, achieving the highest performance with a cosine similarity threshold value of 0.3. This ensured an F1 score of at least 80% across all RDoC domains. The study revealed consistent reductions in all six RDoC domains among PTSD patients after psychotherapy. We found that 60.6% of PTSD women have at least one abnormal instance of the six RDoC domains as compared to PTSD men (51.3%), with 45.1% of PTSD women with higher levels of sensorimotor disturbances compared to men (41.3%). We also found that 57.3% of PTSD patients have at least one abnormal instance of the six RDoC domains based on our records. Also, veterans had the higher abnormalities of negative and positive valence systems (60% and 51.9% of veterans respectively) compared to non-veterans (59.1% and 49.2% respectively). The domains following first diagnoses of PTSD were associated with heightened cue reactivity to trauma, suicide, alcohol, and substance consumption. CONCLUSIONS: The findings provide initial insights into RDoC functioning in different populations and disease trajectories. Natural language processing proves valuable for capturing real-time, context dependent RDoC instances from extensive clinical notes.

Enhancing post-traumatic stress disorder patient assessment: Leveraging Natural Language Processing for Research of Domain Criteria Identification using electronic medical records.

Background: Extracting research of domain criteria (RDoC) from high-risk populations like those with post-traumatic stress disorder (PTSD) is crucial for positive mental health improvements and policy enhancements. The intricacies of collecting, integrating, and effectively leveraging clinical notes for this purpose introduce complexities. Methods: In our study, we created an NLP workflow to analyze electronic medical record (EMR) data, and identify and extract research of domain criteria using a pre-trained transformer-based natural language model, allmpnet-base-v2. We subsequently built dictionaries from 100,000 clinical notes and analyzed 5.67 million clinical notes from 38,807 PTSD patients from the University of Pittsburgh Medical Center. Subsequently, we showcased the significance of our approach by extracting and visualizing RDoC information in two use cases: (i) across multiple patient populations and (ii) throughout various disease trajectories. Results: The sentence transformer model demonstrated superior F1 macro scores across all RDoC domains, achieving the highest performance with a cosine similarity threshold value of 0.3. This ensured an F1 score of at least 80% across all RDoC domains. The study revealed consistent reductions in all six RDoC domains among PTSD patients after psychotherapy. Women had the highest abnormalities of sensorimotor systems, while veterans had the highest abnormalities of negative and positive valence systems. The domains following first diagnoses of PTSD were associated with heightened cue reactivity to trauma, suicide, alcohol, and substance consumption. Conclusions: The findings provide initial insights into RDoC functioning in different populations and disease trajectories. Natural language processing proves valuable for capturing real-time, context dependent RDoC instances from extensive clinical notes.

DeepBiomarker2: Prediction of Alcohol and Substance Use Disorder Risk in Post-Traumatic Stress Disorder Patients Using Electronic Medical Records and Multiple Social Determinants of Health.

Prediction of high-risk events amongst patients with mental disorders is critical for personalized interventions. We developed DeepBiomarker2 by leveraging deep learning and natural language processing to analyze lab tests, medication use, diagnosis, social determinants of health (SDoH) parameters, and psychotherapy for outcome prediction. To increase the model's interpretability, we further refined our contribution analysis to identify key features by scaling with a factor from a reference feature. We applied DeepBiomarker2 to analyze the EMR data of 38,807 patients from the University of Pittsburgh Medical Center diagnosed with post-traumatic stress disorder (PTSD) to determine their risk of developing alcohol and substance use disorder (ASUD). DeepBiomarker2 predicted whether a PTSD patient would have a diagnosis of ASUD within the following 3 months with an average c-statistic (receiver operating characteristic AUC) of 0.93 and average F1 score, precision, and recall of 0.880, 0.895, and 0.866 in the test sets, respectively. Our study found that the medications clindamycin, enalapril, penicillin, valacyclovir, Xarelto/rivaroxaban, moxifloxacin, and atropine and the SDoH parameters access to psychotherapy, living in zip codes with a high normalized vegetative index, Gini index, and low-income segregation may have potential to reduce the risk of ASUDs in PTSD. In conclusion, the integration of SDoH information, coupled with the refined feature contribution analysis, empowers DeepBiomarker2 to accurately predict ASUD risk. Moreover, the model can further identify potential indicators of increased risk along with medications with beneficial effects.

Prediction of Adverse Events Risk in Patients with Comorbid Post- Traumatic Stress Disorder and Alcohol Use Disorder Using Electronic Medical Records by Deep Learning Models.

Background: Prediction of high-risk events in mental disorder patients is crucial. In our previous study, we developed a deep learning model: DeepBiomarker by using electronic medical records (EMR) to predict suicide related event (SRE) risk in post-traumatic stress disorder (PTSD) patients. Methods: We applied DeepBiomarker2 through data integration of multimodal information: lab test, medication, co-morbidities, and social determinants of health. We analyzed EMRs of 5,565 patients from University of Pittsburgh Medical Center with a diagnosis of PTSD and alcohol use disorder (AUD) on risk of developing an adverse event (opioid use disorder, SREs, depression and death). Results: DeepBiomarker2 predicted whether a PTSD + AUD patient will have a diagnosis of any adverse events (SREs, opioid use disorder, depression, death) within 3 months with area under the receiver operator curve (AUROC) of 0.94. We found piroxicam, vilazodone, dronabinol, tenofovir, suvorexant, empagliflozin, famciclovir, veramyst, amantadine, sulfasalazine, and lamivudine to have potential to reduce risk. Conclusions: DeepBiomarker2 can predict multiple adverse event risk with high accuracy and identify potential risk and beneficial factors. Our results offer suggestions for personalized interventions in a variety of clinical and diverse populations.

Epidemiology, clinical presentation, pathophysiology, and management of long COVID: an update.

The increasing number of coronavirus disease 2019 (COVID-19) infections have highlighted the long-term consequences of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection called long COVID. Although the concept and definition of long COVID are described differently across countries and institutions, there is general agreement that it affects multiple systems, including the immune, respiratory, cardiovascular, gastrointestinal, neuropsychological, musculoskeletal, and other systems. This review aims to provide a synthesis of published epidemiology, symptoms, and risk factors of long COVID. We also summarize potential pathophysiological mechanisms and biomarkers for precise prevention, early diagnosis, and accurate treatment of long COVID. Furthermore, we suggest evidence-based guidelines for the comprehensive evaluation and management of long COVID, involving treatment, health systems, health finance, public attitudes, and international cooperation, which is proposed to improve the treatment strategies, preventive measures, and public health policy making of long COVID.

DeepBiomarker2: Prediction of alcohol and substance use disorder risk in post-traumatic stress disorder patients using electronic medical records and multiple social determinants of health.

Introduction: Prediction of high-risk events amongst patients with mental disorders is critical for personalized interventions. In our previous study, we developed a deep learning-based model, DeepBiomarker by utilizing electronic medical records (EMR) to predict the outcomes of patients with suicide-related events in post-traumatic stress disorder (PTSD) patients. Methods: We improved our deep learning model to develop DeepBiomarker2 through data integration of multimodal information: lab tests, medication use, diagnosis, and social determinants of health (SDoH) parameters (both individual and neighborhood level) from EMR data for outcome prediction. We further refined our contribution analysis for identifying key factors. We applied DeepBiomarker2 to analyze EMR data of 38,807 patients from University of Pittsburgh Medical Center diagnosed with PTSD to determine their risk of developing alcohol and substance use disorder (ASUD). Results: DeepBiomarker2 predicted whether a PTSD patient will have a diagnosis of ASUD within the following 3 months with a c-statistic (receiver operating characteristic AUC) of 0·93. We used contribution analysis technology to identify key lab tests, medication use and diagnosis for ASUD prediction. These identified factors imply that the regulation of the energy metabolism, blood circulation, inflammation, and microbiome is involved in shaping the pathophysiological pathways promoting ASUD risks in PTSD patients. Our study found protective medications such as oxybutynin, magnesium oxide, clindamycin, cetirizine, montelukast and venlafaxine all have a potential to reduce risk of ASUDs. Discussion: DeepBiomarker2 can predict ASUD risk with high accuracy and can further identify potential risk factors along with medications with beneficial effects. We believe that our approach will help in personalized interventions of PTSD for a variety of clinical scenarios.

Review article: Effective management of opioid withdrawal symptoms: A gateway to opioid dependence treatment.

BACKGROUND AND OBJECTIVES: The opioid crisis has taken an immense toll in the United States. On average, five lives are lost to an opioid overdose every hour of the day; estimated costs associated with opioid misuse exceed $500 billion annually. Illicit opioid discontinuation is the first step in the treatment of opioid use disorder (OUD), and transition to an opioid agonist may initiate treatment. However, discontinuation to abstinence from either OUD directly or following agonist treatment results in severely distressing opioid withdrawal symptoms (OWS). METHODS: This review evaluated studies on the etiology, burden, and management of OWS. RESULTS: Noradrenergic hyperactivity generates many OWS. These OWS can cause patients to relapse during early opioid discontinuation. While agonist therapies are generally first-line for moderate or severe OUD and reduce OWS, prescribing restrictions can limit their availability. DISCUSSION AND CONCLUSIONS: Non-opioid medications to treat OWS provides a gateway into long-term treatment with naltrexone or psychosocial therapies. For opioid dependent patients without OUD, non-opioid treatments like α-2 adrenergic agonists can facilitate opioid tapering. SCIENTIFIC SIGNIFICANCE: For the millions who are physically dependent on opioids, new treatments for OWS can enhance recovery from OUD and prevent relapse. (© 2019 The Authors. The American Journal on Addictions Published by Wiley Periodicals, Inc.;XX:1-8).

Buprenorphine Prescribing: To Expand or Not to Expand.

As a result of the prescription opioid epidemic in the United States, there has been an increasing need for effective treatment interventions, both pharmacological and nonpharmacological. Buprenorphine has emerged as a critical component of the treatment of opioid use disorder, yet its adoption has not been without some concerns. This article first reviews the pharmacology, clinical use, and US legislative action related to buprenorphine, followed by a discussion of the misuse and diversion of buprenorphine in the United States as well as internationally. We then explore the impact of buprenorphine abuse as well as discussing strategies for its reduction, including changes in policy, prescription and pharmacy monitoring, and continuing medical education for guiding and improving clinical practice.

Smoking reduction and quality of life in chronic patients with schizophrenia in a Chinese population--A pilot study.

BACKGROUND AND OBJECTIVES: Tobacco use is a significant public health issue on a global scale. Prevalence of daily tobacco smoking for men in China is much higher than in the United States. Although prevailing literature suggests a negative relationship between smoking and quality of life, this pilot study sought to evaluate whether smoking reduction/cessation impacted on the perception of quality of life in an in-patient population in China. METHODS: Twenty Chinese patients meeting DSM-IV criteria for schizophrenia were recruited from Beijing Hui-Long-Guan Hospital, an in-patient facility in Beijing, China, for participation in this 4-week study. Seventeen participants with schizophrenia completed the study and were included in the final analysis. Cigarette consumption was recorded daily and the World Health Organization Quality of Life-BREF (WHOQOL-BREF) was completed at baseline and at week 4. The relationships between smoking and perceived quality of life were evaluated using correlations between changes in WHOQOL-BREF and changes in cigarettes consumed as measured from baseline to week 4. RESULTS: We found an increase in perceived quality of life in the social relationships domain with increased cigarette consumption in contrast to a decrease in this domain with decreased consumption. However, decreased cigarette consumption was associated with an increase in the psychological domain compared to the social domain. CONCLUSIONS AND SCIENTIFIC SIGNIFICANCE: These associations suggest a need for interventions to improve the social relationship perceptions with any successful reduction in cigarette consumption among Chinese schizophrenics in order to match their perceived psychological improvement.

Gender differences in socio-demographic and clinical characteristics of methamphetamine inpatients in a Chinese population.

BACKGROUND: Over the past two decades, China has experienced a dramatic increase in methamphetamine (MA) abuse. This study examined gender-specific socio-demographic and clinical characteristics of MA use among Han Chinese, which has previously received little systematic study. METHODS: This analysis described MA-related socio-demographic and clinical characteristics in a broad cross-sectional sample (n=1464; male/female=1185/279), and examined differences between males and females in MA use history, MA initiation, MA-related subjective feelings and behaviors, and withdrawal symptoms. RESULTS: Most MA abusers (about 72%) were young (in their 20s or 30s), with women being 5 years younger than men on average. More males (33.2%) were married than females (21.9%). The average body mass index (BMI) was significantly lower in this MA abuser sample compared to the age-matched healthy controls. Moreover, the BMI of females was significantly lower than that of males. The laboratory tests showed that the blood levels of glucose, cholesterol and triglyceride were all significantly higher in males than females. Females used MA at a younger age compared to males. The most frequent route of MA use was smoking (90.9%). Males were more likely to use another drug, and more likely to be hospitalized. However, many characteristics and behaviors of MA use are similar for males and females, including the route, the dose and duration of MA use, and relapse status. CONCLUSION: Although there were some male-female similarities in MA use parameters, significant differences do exist that may have implications for gender-specific research as well as for prevention and treatment strategies.

Performance-based assessment of functional skills in severe mental illness: results of a large-scale study in China.

Performance-based assessments of everyday living skills have been shown to be highly correlated with cognitive functioning in schizophrenia and bipolar disorder, as well as being predictive of deficits in real-world outcomes such as independent living and employment. In this study, we expand our assessments of impairments in everyday living skills to China, evaluating people with schizophrenia, bipolar disorder, and major depression, and comparing their performance to that of healthy controls. Samples of people with schizophrenia (n = 272), bipolar disorder (n = 61), major depression (n = 50), and healthy controls (n = 284) were examined with the Chinese version of the UCSD performance-based assessment, brief version (UPSA-B). Performance was compared across the groups and the association between age, gender, educational attainment, marital status, and UPSA-B scores was evaluated. When the performance on the UPSA was compared across the groups, with education as a covariate, significant effects of both diagnosis (F = 86.3, p < .001) and education were found (F = 228.3, p < .001). Sex and age did not contribute significantly when age and education were considered. Post-hoc comparisons revealed that total UPSA-B scores were lowest in the schizophrenia patients, followed by the patients with major depression. Patients with bipolar disorder did not differ from the healthy comparison subjects on overall performance. Scores for all groups were lower than previously reported in western samples (e.g., HC mean = 64). While diagnostic differences in UPSA-B scores are similar to those previously seen in western samples, the education effect is considerably more substantial. These data suggest that in developing countries educational attainment may be strongly associated with levels of adaptive outcomes and the utilization and interpretation of functional capacity measures be adjusted accordingly.

Therapeutic options and challenges for substances of abuse.

Addiction to substances continues to be a significant public health concern in the United States. The following review of current pharmacological treatments discusses a range of substances: nicotine, alcohol, cocaine, and opioids. The goal is to provide an overview of currently available and new pharmacological treatments for substance use disorders, while also addressing the pharmacotherapeutic challenges remaining. The significant advances in pharmacotherapy have had limited utilization, however. For example, naltrexone for alcoholism is infrequently prescribed, buprenorphine for opiates still has relatively few qualified prescribers, and stimulants have no Food and Drug Administration-approved pharmacotherapy. These pharmacotherapies are needed, with the rate of even the relatively uncommon abuse of opiates now rising sharply.

Choosing a behavioral therapy platform for pharmacotherapy of substance users.

Behavioral therapy platforms have become virtual requirements in pharmacotherapy trials due to their utility in reducing noise variability, preventing differential medication adherence and protocol attrition, enhancing statistical power and addressing ethical issues in placebo-controlled trials. Selecting an appropriate behavioral platform for a particular trial requires study-specific tailoring, taking into account both the stage of development of the medication being evaluated, as well as the specific strengths and weaknesses of a broad array of available empirically supported behavioral therapies and the range of their possible targets (e.g., enhancing medication adherence, preventing attrition, addressing co-morbid problems, fostering abstinence, and targeting specific weaknesses of the pharmacologic agent). Choosing a suitable behavioral platform also requires consideration of the characteristics of the population to be treated, stage of scientific knowledge regarding the medication's effects, appropriate balance of internal and external validity, and consideration of potential ceiling effects. Available manualized behavioral treatments are reviewed, noting their strengths and limitations as behavioral therapy platforms for pharmacotherapy trials and as potential concomitant therapies in clinical practice.

Measuring alcohol consumption among older adults: a comparison of available methods.

We determined levels of agreement and concordance between five alcohol measures among older veterans from a VA primary care clinic (N=303) and community-dwelling Medicare beneficiaries (N=511). The individuals recruited for the study had not been previously identified for alcohol use. Quantity-frequency questions, a binge drinking question, the Alcohol Use Disorders Identification Test, CAGE, and questions on lifetime consumption were administered to all current drinkers. To assess agreement between the measures, we conducted pairwise comparisons (eg, QF vs. CAGE) by determining the percentage of participants who screened positive on both measures and calculated kappa values to determine concordance. Only modest levels of agreement and concordance were found between the measures. The use of any individual measure may fail to detect many older adults with important alcohol exposures. Until optimal methods for measuring alcohol exposures in elderly adults are defined, we suggest that a combination of measures be used when obtaining alcohol histories in older persons.