Dataset Information

Predictive Modelling of Susceptibility to Substance Abuse, Mortality and Drug-Drug Interactions in Opioid Patients.

ABSTRACT: Objective: Opioids are a class of drugs that are known for their use as pain relievers. They bind to opioid receptors on nerve cells in the brain and the nervous system to mitigate pain. Addiction is one of the chronic and primary adverse events of prolonged usage of opioids. They may also cause psychological disorders, muscle pain, depression, anxiety attacks etc. In this study, we present a collection of predictive models to identify patients at risk of opioid abuse and mortality by using their prescription histories. Also, we discover particularly threatening drug-drug interactions in the context of opioid usage. Methods and Materials: Using a publicly available dataset from MIMIC-III, two models were trained, Logistic Regression with L2 regularization (baseline) and Extreme Gradient Boosting (enhanced model), to classify the patients of interest into two categories based on their susceptibility to opioid abuse. We've also used K-Means clustering, an unsupervised algorithm, to explore drug-drug interactions that might be of concern. Results: The baseline model for classifying patients susceptible to opioid abuse has an F1 score of 76.64% (accuracy 77.16%) while the enhanced model has an F1 score of 94.45% (accuracy 94.35%). These models can be used as a preliminary step towards inferring the causal effect of opioid usage and can help monitor the prescription practices to minimize the opioid abuse. Discussion and Conclusion: Results suggest that the enhanced model provides a promising approach in preemptive identification of patients at risk for opioid abuse. By discovering and correlating the patterns contributing to opioid overdose or abuse among a variety of patients, machine learning models can be used as an efficient tool to help uncover the existing gaps and/or fraudulent practices in prescription writing. To quote an example of one such incidental finding, our study discovered that insulin might possibly be interacting with opioids in an unfavourable way leading to complications in diabetic patients. This indicates that diabetic patients under long term opioid usage might need to take increased amounts of insulin to make it more effective. This observation backs up prior research studies done on a similar aspect. To increase the translational value of our work, the predictive models and the associated software code are made available under the MIT License.

SUBMITTER: Vunikili R

PROVIDER: S-EPMC8702828 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundThe abuse of psychogenic drugs can lead to multiple health-related problems. Genetic and environmental vulnerabilities are factors in the emergence of substance use disorders. Empirical evidence regarding the gene-environment interaction in substance use is mixed. Summaries of the latest findings from a candidate gene approach will be useful for revealing the significance of particular gene contributions. Thus, we aim to identify different gene-environment interactions in patterns of substance use and investigate whether any effects trend notably across different genders and races.MethodsWe reviewed published studies, until March 1, 2022, on substance use for candidate gene-environment interaction. Basic demographics of the included studies, target genes, environmental factors, main findings, patterns of gene-environment interaction, and other relevant information were collected and summarized.ResultsAmong a total of 44 studies, 38 demonstrated at least one significant interaction effect. About 61.5% of studies on the 5-HTTLPR gene, 100% on the MAOA gene, 42.9% on the DRD2 gene, 50% on the DRD4 gene, 50% on the DAT gene, 80% on the CRHR1 gene, 100% on the OPRM1 gene, 100% on the GABRA1 gene, and 50% on the CHRNA gene had a significant gene-environment interaction effect. The diathesis-stress model represents a dominant interaction pattern (89.5%) in the studies with a significant interaction effect; the remaining significant effect on substance use is found in the differential susceptibility model. The social push and swing model were not reported in the included studies.ConclusionThe gene-environment interaction research on substance use behavior is methodologically multidimensional, which causes difficulty in conducting pooled analysis, or stated differently-making it hard to identify single sources of significant influence over maladaptive patterns of drug taking. In decreasing the heterogeneity and facilitating future pooled analysis, researchers must (1) replicate the existing studies with consistent study designs and measures, (2) conduct power calculations to report gene-environment correlations, (3) control for covariates, and (4) generate theory-based hypotheses with factorial based experiments when designing future studies.

Project description:BackgroundForensic psychiatric patients are often diagnosed with psychiatric disorders characterized by high levels of impulsivity as well as comorbid substance use disorders (SUD). The combination of psychiatric disorders and SUD increases the risk of future violence. Chronic substance abuse can lead to a structural state of disinhibition, resulting in more drug taking and eventually loss of control over drug intake. When treating SUD, it is crucial to address high levels of impulsivity and lack of inhibitory control.ObjectiveThis study set out to investigate the effects of a theta/sensorimotor rhythm (SMR) neurofeedback training protocol on levels of impulsivity, levels of drug craving, and actual drug intake in a population of forensic psychiatric patients with a diagnosis of SUD.MethodsA total of 21 participants received 20 sessions of theta/SMR neurofeedback training in combination with treatment-as-usual (TAU). Results of the intervention were compared with results from 21 participants who received TAU only.ResultsSMR magnitude showed a significant (P=.02) increase post training for patients in the neurofeedback training group, whereas theta magnitude did not change (P=.71). Levels of drug craving as well as scores on the motor subscale of the Barratt Impulsivity Scale-11 decreased equally for patients in the neurofeedback training group and the TAU group. Other measures of impulsivity as well as drug intake did not change posttreatment (P>.05). Therefore, neurofeedback+TAU was not more effective than TAU only.ConclusionsThis study demonstrated evidence that forensic psychiatric patients are able to increase SMR magnitude over the course of neurofeedback training. However, at the group level, the increase in SMR activity was not related to any of the included impulsivity or drug craving measures. Further research should focus on which patients will be able to benefit from neurofeedback training at an early stage of the employed training sessions.Trial registrationDutch National Trial Register: NTR5386; http://www.trialregister.nl/trialreg/admin/rctview.asp?TC=5386 (Archived by WebCite at http://www.webcitation.org/6nXLQuoLl).

Project description:BackgroundSubstance abuse has an impact on various cognitive domains, including memory. Even though this impact has been extensively examined across different subdomains, false memory has been sparsely studied. This systematic review and meta-analysis seek to synthesize the current scientific data concerning false memory formation in individuals with a history of substance abuse.MethodsPubMed, Scopus, the Cochrane Library, Web of Science, and PsycINFO were searched to identify all experimental and observational studies in English, Portuguese, and Spanish. Studies were then examined by four independent reviewers and, if they met the inclusion criteria, assessed for their quality. The Cochrane Risk of Bias Tool for randomized controlled trials (RCT) and the Joanna Briggs Institute (JBI) critical appraisal checklists for quasi-experimental and analytic cross-sectional studies were used to assess the risk of bias.ResultsFrom the 443 screened studies, 27 (and two more from other sources) were considered eligible for full-text review. A final 18 studies were included in the present review. Of these, 10 were conducted with alcoholics or heavy drinkers, four focused on ecstasy/polydrug users, three were done with cannabis users and one focused on methadone maintenance patients with current cocaine dependence. Regarding false memory type, 15 studies focused on false recognition/recall, and three on provoked confabulation.ConclusionsNone but one of the studies considering false recognition/recall of critical lures found any significant differences between individuals with a history of substance abuse and healthy controls. However, most of the studies taking into account false recognition/recall of related and unrelated events found that individuals with a history of substance abuse showed significantly higher rates of false memories than controls. Future research should continue to consider different types of false memories as well as their potential association with relevant clinical variables.Systematic review registrationhttps://www.crd.york.ac.uk/prospero/display_record.php?RecordID=266503, identifier: CRD42021266503.

Project description:BackgroundWarfarin is effective in preventing thromboembolic events, but concerns exist regarding its use in patients with substance abuse.ObjectiveIdentify which patients with substance abuse who receive warfarin are at risk for poor outcomes.DesignRetrospective cohort study. Diagnostic codes, lab values, and other factors were examined to identify risk of adverse outcomes.PatientsVeterans AffaiRs Study to Improve Anticoagulation (VARIA) database of 103,897 patients receiving warfarin across 100 sites.Main measuresOutcomes included percent time in therapeutic range (TTR), a measure of anticoagulation control, and major hemorrhagic events by ICD-9 codes.ResultsNonusers had a higher mean TTR (62 %) than those abusing alcohol (53 %), drugs (50 %), or both (44 %, p?<?0.001). Among alcohol abusers, an increasing ratio of the serum hepatic transaminases aspartate aminotransferase/alanine aminotransferase (AST:ALT) correlated with inferior anticoagulation control; normal AST:ALT?? 1.5 predicted relatively modest decline in TTR (54 %, p?<?0.001), while elevated ratios (AST:ALT 1.50-2.0 and > 2.0) predicted progressively poorer anticoagulation control (49 % and 44 %, p?<?0.001 compared to nonusers). Age-adjusted hazard ratio for major hemorrhage was 1.93 in drug and 1.37 in alcohol abuse (p?<?0.001 compared to nonusers), and remained significant after also controlling for anticoagulation control and other bleeding risk factors (1.69 p?<?0.001 and 1.22 p?=?0.003). Among alcohol abusers, elevated AST:ALT >2.0 corresponded to more than three times the hemorrhages (HR 3.02, p?<?0.001 compared to nonusers), while a normal ratio AST:ALT ? 1.5 predicted a rate similar to nonusers (HR 1.19, p?<?0.05).ConclusionsAnticoagulation control is particularly poor in patients with substance abuse. Major hemorrhages are more common in both alcohol and drug users. Among alcohol abusers, the ratio of AST/ALT holds promise for identifying those at highest risk for adverse events.

Dataset Information

Predictive Modelling of Susceptibility to Substance Abuse, Mortality and Drug-Drug Interactions in Opioid Patients.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets