Usage Events Research Articles

Evaluating the feasibility of a Mobile Audio Companion (Elly) to reduce anxiety and to improve patient-reported outcomes among patients with cancer.

11120 Background: Anxiety is one of the most commonly reported symptoms among cancer patients, and is a known risk factor for reduced health-related quality of life and other negative patient-reported outcomes (PROs). Mobile health applications (apps) are tools that can be used flexibly in patients’ preferred environments with the potential to reduce anxiety and improve quality of life. The current study sought to examine the feasibility and preliminary efficacy of a mobile audio companion app, Elly, to improve anxiety and other critical PROs among cancer patients. Methods: This was a single-site, single-arm pilot study conducted over 6-months. Adults 18 years or older with a diagnosis of cancer of any type, received treatment within 6 months of consent, and had access to an iPhone were eligible. The primary outcome was change in mean anxiety t-scores (NIH PROMIS anxiety short form (8a)) administered electronically via REDCap at baseline, 1 (primary timepoint), 3, and 6 months. Secondary outcomes included changes in NIH PROMIS depression, loneliness, social support, and perceived stress scores from baseline to 6 months. Changes in NIH PROMIS T-scores at each time point were calculated and analyzed using paired t-tests. Results: Of 53 patients who consented for the study, 38 patients (72%) were included in the analysis ( Mage: 52.8 years, SD: 11.7, 84.2% female). Most participants reported a history of breast cancer (50%), followed by ovarian (16%), hematologic (11%), and other cancer types (24%) Phone incompatibility and technological challenges were the most common reasons for screen-fails (n=15). All participants downloaded and used the app at least once, with an average of 154 app usage events across participants recorded over the study period (median: 64, range 11-1186 times). The mean baseline anxiety T-score was 63.5 (SD: 2.1). Anxiety levels decreased among participants over time, with a mean difference in anxiety T-scores of -1.9 at 30 days, -3.3 at 90 days, and -3.0 at 6 months from baseline, which was statistically (p<0.05) and clinically significant. Participants also reported improvements in perceived stress, depression symptomatology, social support, and loneliness, although statistical significance was not reached. Conclusions: Mobile apps, such as Elly, may feasibly be used to support patient-reported anxiety as well as other PROs, especially among those who present with high levels of anxiety at baseline. Efforts to increase access to mobile health apps and provide technical support for patients should be made increase equity across the cancer survivor population. Larger, multi-center randomized controlled trials are needed to build on these findings.

Investigating Rhythmicity in App Usage to Predict Depressive Symptoms: Protocol for Personalized Framework Development and Validation Through a Countrywide Study.

Understanding a student's depressive symptoms could facilitate significantly more precise diagnosis and treatment. However, few studies have focused on depressive symptom prediction through unobtrusive systems, and these studies are limited by small sample sizes, low performance, and the requirement for higher resources. In addition, research has not explored whether statistically significant rhythms based on different app usage behavioral markers (eg, app usage sessions) exist that could be useful in finding subtle differences to predict with higher accuracy like the models based on rhythms of physiological data. The main objective of this study is to explore whether there exist statistically significant rhythms in resource-insensitive app usage behavioral markers and predict depressive symptoms through these marker-based rhythmic features. Another objective of this study is to understand whether there is a potential link between rhythmic features and depressive symptoms. Through a countrywide study, we collected 2952 students' raw app usage behavioral data and responses to the 9 depressive symptoms in the 9-item Patient Health Questionnaire (PHQ-9). The behavioral data were retrieved through our developed app, which was previously used in our pilot studies in Bangladesh on different research problems. To explore whether there is a rhythm based on app usage data, we will conduct a zero-amplitude test. In addition, we will develop a cosinor model for each participant to extract rhythmic parameters (eg, acrophase). In addition, to obtain a comprehensive picture of the rhythms, we will explore nonparametric rhythmic features (eg, interdaily stability). Furthermore, we will conduct regression analysis to understand the association of rhythmic features with depressive symptoms. Finally, we will develop a personalized multitask learning (MTL) framework to predict symptoms through rhythmic features. After applying inclusion criteria (eg, having app usage data of at least 2 days to explore rhythmicity), we kept the data of 2902 (98.31%) students for analysis, with 24.48 million app usage events, and 7 days' app usage of 2849 (98.17%) students. The students are from all 8 divisions of Bangladesh, both public and private universities (19 different universities and 52 different departments). We are analyzing the data and will publish the findings in a peer-reviewed publication. Having an in-depth understanding of app usage rhythms and their connection with depressive symptoms through a countrywide study can significantly help health care professionals and researchers better understand depressed students and may create possibilities for using app usage-based rhythms for intervention. In addition, the MTL framework based on app usage rhythmic features may more accurately predict depressive symptoms due to the rhythms' capability to find subtle differences. DERR1-10.2196/51540.

MicroRNA Transcriptomes Reveal Prevalence of Rare and Species-Specific Arm Switching Events During Zebrafish Ontogenesis.

In metazoans, microRNAs (miRNAs) are essential regulators of gene expression, affecting critical cellular processes from differentiation and proliferation, to homeostasis. During miRNA biogenesis, the miRNA strand that loads onto the RNA-induced Silencing Complex (RISC) can vary, leading to changes in gene targeting and modulation of biological pathways. To investigate the impact of these "arm switching" events on gene regulation, we analyzed a diverse range of tissues and developmental stages in zebrafish by comparing 5p and 3p arms accumulation dynamics between embryonic developmental stages, adult tissues, and sexes. We also compared variable arm usage patterns observed in zebrafish to other vertebrates including arm switching data from fish, birds, and mammals. Our comprehensive analysis revealed that variable arm usage events predominantly take place during embryonic development. It is also noteworthy that isomiR occurrence correlates to changes in arm selection evidencing an important role of microRNA distinct isoforms in reinforcing and modifying gene regulation by promoting dynamics switches on miRNA 5p and 3p arms accumulation. Our results shed new light on the emergence and coordination of gene expression regulation and pave the way for future investigations in this field.

Trusting information systems in everyday work events – effects on cognitive resources, performance, and well-being

In today’s data-intensive work environments, information systems are crucial for supporting workers. However, workers often do not rely on these systems but resort to workarounds. We argue that trust is essential for workers’ reliance on information systems, positively affecting workers’ cognitive resources, performance, and well-being. Moreover, we argue that the organisational context (accountability, distractions) and user-related factors qualify trust-outcome associations by affecting workers’ trust calibration. In a preregistered study, we asked N = 291 employed users of information systems to re-experience prior everyday usage events (event reconstruction method) and assess event-specific trust in the system, work outcomes, and context conditions. Results confirmed the assumed association between trust in the information system and workers’ ratings of both performance and well-being. Moreover, workers’ technology competence and need for cognition – but not contextual conditions – qualified trust-outcome associations. Our results offer specific suggestions for achieving successful use of information systems at work.

Usage events and constructional knowledge: A study of two variants of the introductory-it construction – CORRIGENDUM

Usage events and constructional knowledge: A study of two variants of the introductory-it construction – CORRIGENDUM

Usage events and constructional knowledge: A study of two variants of the introductory-it construction

AbstractUsage-based theories hold that mental representation of language is shaped by a lifetime of usage. Both input to which first language (L1) and second language (L2) users are exposed and their own language production affect their construction learning and entrenchment. The present study investigates L2 users’ knowledge of two introductory-it variants, Adj-that (e.g., it is clear that …) and Adj-to (e.g., it is difficult to …). We probed the extent to which adjective–variant associations in an academic section of COCA and L2 users’ engagement with academic writing affected learners’ generation of adjectives distinctively attracted to the two variants. An analysis of cue-outcome contingency was conducted to establish adjective–variant associations, and an elicitation task was carried out, probing L2 users’ ability to generate adjectives when prompted with the variants (e.g., it is [blank] to). The participants were 84 graduate students in the United States, 44 from L1 English and 40 from L1 Thai backgrounds. The results indicated that the adjective–variant associations predicted L2 users’ generation of adjectives. However, academic writing engagement did not affect learners’ performance. The findings suggest that statistical information in the input affects L2 users’ constructional representation.

The status of resistance to phosphine in common bruchid species infesting edible stored pulses in India

Bruchids or pulses beetles or bean weevils (Coleoptera: Chrysomelidae: Bruchinae) are serious pests of stored pulses, and the only curative chemical option approved for use in India is fumigation with phosphine. Relative phosphine resistance was ascertained in 79 bruchid populations from 74 diverse locations of India, in comparison to their susceptible laboratory conspecifics. The populations' response to phosphine was varied vis-à-vis species, locations, and storage conditions. Overall, the populations recorded the relative resistance ratio of 1.1–4.5 folds over their susceptible laboratory conspecifics, however the LC99 values indicated populations are still susceptible to recommended phosphine dosage by Central Insecticide Board and Registration Committee. The relative resistance levels in decreasing order for species was Callosobruchus maculatus > C. analis > C. chinensis. The populations from small-scale storage conditions exhibited relatively lower level of resistance than those from large-scale storage conditions. The relative resistance was varied between populations representing different states and regions of India. The populations from places where pulses are grown, stored and protected with phosphine on a large-scale and regular basis showed higher relative resistance. When compared to species infesting stored cereals, the relative resistance levels recorded in bruchid populations was rather low; however, the observed levels warn of further escalation to an alarming level in the events of inappropriate and indiscriminate phosphine usage. Consequently, the Indian pulses storages fully reliant on this fumigant may face a significant bruchid threat in the near future. Therefore, the current study provides a valuable source of information on phosphine resistance status in bruchids infesting edible stored pulses across India.

A panel of TDP-43-regulated splicing events verifies loss of TDP-43 function in amyotrophic lateral sclerosis brain tissue

TDP-43 dysfunction is a molecular hallmark of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). A major hypothesis of TDP-43 dysfunction in disease is the loss of normal nuclear function, resulting in impaired RNA regulation and the emergence of cryptic exons. Cryptic exons and differential exon usage are emerging as promising markers of lost TDP-43 function in addition to revealing biological pathways involved in neurodegeneration in ALS/FTD. In this brief report, we identified markers of TDP-43 loss of function by depleting TARDBP from post-mortem human brain pericytes, a manipulable in vitro primary human brain cell model, and identifying differential exon usage events with bulk RNA-sequencing analysis. We present these data in an interactive database (https://www.scotterlab.auckland.ac.nz/research-themes/tdp43-lof-db-v2/) together with seven other TDP-43-depletion datasets we meta-analysed previously, for user analysis of differential expression and splicing signatures. Differential exon usage events that were validated by qPCR were then compiled into a ‘differential exon usage panel’ with other well-established TDP-43 loss-of-function exon markers. This differential exon usage panel was investigated in ALS and control motor cortex tissue to verify whether, and to what extent, TDP-43 loss of function occurs in ALS. We find that profiles of TDP-43-regulated cryptic exons, changed exon usage and changed 3′ UTR usage discriminate ALS brain tissue from controls, verifying that TDP-43 loss of function occurs in ALS. We propose that TDP-43-regulated splicing events that occur in brain tissue will have promise as predictors of disease.

ATPP : A Mobile App Prediction System Based on Deep Marked Temporal Point Processes

Predicting the next application (app) a user will open is essential for improving the user experience, e.g., app pre-loading and app recommendation. Unlike previous solutions that only predict which app the user will open, this article predicts both the next app and the time to open it. Time prediction is essential to avoid loading the next app too early and consuming unnecessary resources on smartphones. To predict the next app and open time jointly, we model the app usage sequence as a marked temporal point process (MTPP), whose conditional intensity function can capture the probability of a new app usage event. We develop a novel data-driven MTPP-based app prediction system, named ATPP (App Temporal Point Process), which adopts a recurrent neural network architecture to learn the MTPP conditional intensity function for app prediction. ATPP adopts a set of techniques to incorporate the unique features of app prediction in our RNN architecture, including learning the correlated usage behavior of different apps by representation learning, the temporal dependency of app usage by an attention mechanism, and the location-related app usage behavior by feature extraction and fusion layer. We conduct extensive experiments on a large-scale anonymized app usage dataset to verify ATPP’s effectiveness.

Unveiling the Role of Dynamic Alternative Splicing Modulation After Infestation with Sea Lice (Caligus rogercresseyi) in Atlantic Salmon.

Sea lice are pathogenic marine ectoparasite copepods that represent a severe risk to the worldwide salmon industry. Several transcriptomic investigations have characterized the regulation of gene expression response of Atlantic salmon to sea lice infestation. These studies have focused on the levels of transcript, overlooking the potentially relevant role of alternative splicing (AS), which corresponds to an essential control mechanism of gene expression through RNA processing. In the present study, we performed a genome-wide bioinformatics characterization of differential AS event dynamics in control and infested C. rogercresseyi Atlantic salmon and in resistant and susceptible phenotypes. We identified a significant rise of alternative splicing events and AS genes after infestation and 176 differential alternative splicing events (DASE) from 133 genes. In addition, a higher number of DASE and AS genes were observed among resistant and susceptible phenotypes. Functional annotation of AS genes shows several terms and pathways associated with behavior, RNA splicing, immune response, and RNA binding. Furthermore, three protein-coding genes were identified undergoing differential transcript usage events, among resistant and susceptible phenotypes. Our findings support AS performing a relevant regulatory role in the response of salmonids to sea lice infestation.

Mapping musical dynamics in space. A qualitative analysis of conductors' movements in orchestra rehearsals

In this contribution, we examine the way in which orchestra conductors use the space around them to convey aspects of musical dynamics. In music,dynamicsrefers to the intensity of volume of notes and sounds and its interpretation is highly context-bound. We approach dynamics as a phenomenon of emergent and construed meaning in interaction, induced by the music score and the interpretationin situby musicians and the conductor. Conductors' movement-based instructions on dynamics result in highly complex usage events. This study aims at disentangling these instances by asking how conductors move and use the space around them to instruct on (un)desired aspects of musical dynamics, zooming in on movement direction as a central formal feature. We find ourselves at the crossroads of cognitive and interactional linguistics, aligning with existing studies on the interactional and contextually embedded nature of music interaction. From a cognitive linguistic perspective, this endeavor translates as the identification of the construal mechanisms (metaphor, specificity and viewpoint) that underlie and therefore motivate movement directions in the specific instances under examination. The analysis is based on 10h of video data from a corpus recorded during rehearsals of five conductors instructing their respective orchestras in Dutch. Our data reveal that conductors use different movement patterns, some of which appear to involve opposite movement directions for expressing a similar music dynamical aspect, e.g., depending on the usage event, a vertical upward movement can mean both a request for playing louder and softer. By taking into account different construal mechanisms, we are able to provide an encompassing multimodal analysis, in which these allegedly deviating oppositional movements appear as consistently motivated (metaphorical) expressions, which profile a similar target concept involving different viewpoints.

Communicative Competence Fostered in a Nested EFL Learning Ecology: Technology-Enhanced Learning in the Chinese Context

This study investigates the self-perceived effectiveness of a nested EFL learning model designed to create an affordance-rich learning ecology to help EFL learners realize the co-evolution of vocabulary knowledge and communicative competence. Adopting an action research approach, this study examined the predicament college students encountered under the CLT paradigm in the Chinese context, and the learning difficulties were identified. A nested EFL learning model was designed to address the issue by creating an effective learning environment in which micro, meso and exo systems interacted with each other to provide multi-mode usage events for the EFL learners’ language development. Quantitative data was collected from two online surveys (N=91) and analyzed with IBM SPSS 26, and qualitative data was elicited from students’ reflective accounts. The results revealed that the model positively affected students' language development regarding their vocabulary acquisition, listening and speaking proficiency. Other benefits derived from the learning experience outside the classroom, represented by an innovative WeCo (We Connect) reports project based on WeChat, are reflected in students’ growing self-confidence as EFL learners and their awakening consciousness to learn English by using the language. Some constructive suggestions were proposed to improve the model. The design in this study provides a solution to the learning difficulties perplexing most EFL learners. It sheds light on how teacher-guided self-regulated English learning beyond the classroom can be augmented by modern technologies to develop communicative competence accompanied by the growth of productive vocabulary knowledge.

Transcriptional targets of amyotrophic lateral sclerosis/frontotemporal dementia protein TDP-43 - meta-analysis and interactive graphical database.

ABSTRACTTDP-43 proteinopathy is the major pathology in amyotrophic lateral sclerosis (ALS) and tau-negative frontotemporal dementia (FTD). Mounting evidence implicates loss of normal TDP-43 RNA-processing function as a key pathomechanism. However, the RNA targets of TDP-43 differ by report, and have never been formally collated or compared between models and disease, hampering understanding of TDP-43 function. Here, we conducted re-analysis and meta-analysis of publicly available RNA-sequencing datasets from six TDP-43-knockdown models, and TDP-43-immunonegative neuronal nuclei from ALS/FTD brain, to identify differentially expressed genes (DEGs) and differential exon usage (DEU) events. There was little overlap in DEGs between knockdown models, but PFKP, STMN2, CFP, KIAA1324 and TRHDE were common targets and were also differentially expressed in TDP-43-immunonegative neurons. DEG enrichment analysis revealed diverse biological pathways including immune and synaptic functions. Common DEU events in human datasets included well-known targets POLDIP3 and STMN2, and novel targets EXD3, MMAB, DLG5 and GOSR2. Our interactive database (https://www.scotterlab.auckland.ac.nz/research-themes/tdp43-lof-db/) allows further exploration of TDP-43 DEG and DEU targets. Together, these data identify TDP-43 targets that can be exploited therapeutically or used to validate loss-of-function processes.This article has an associated First Person interview with the first author of the paper.

Exploring differential exon usage via short- and long-read RNA sequencing strategies.

Alternative splicing produces various mRNAs, and thereby various protein products, from one gene, impacting a wide range of cellular activities. However, accurate reconstruction and quantification of full-length transcripts using short-reads is limited, due to their length. Long-reads sequencing technologies may provide a solution by sequencing full-length transcripts. We explored the use of both Illumina short-reads and two long Oxford Nanopore Technology (cDNA and Direct RNA) RNA-Seq reads for detecting global differential splicing during mouse embryonic stem cell differentiation, applying several bioinformatics strategies: gene-based, isoform-based and exon-based. We detected the strongest similarity among the sequencing platforms at the gene level compared to exon-based and isoform-based. Furthermore, the exon-based strategy discovered many differential exon usage (DEU) events, mostly in a platform-dependent manner and in non-differentially expressed genes. Thus, the platforms complemented each other in the ability to detect DEUs (i.e. long-reads exhibited an advantage in detecting DEUs at the UTRs, and short-reads detected more DEUs). Exons within 20 genes, detected in one or more platforms, were here validated by PCR, including key differentiation genes, such as Mdb3 and Aplp1. We provide an important analysis resource for discovering transcriptome changes during stem cell differentiation and insights for analysing such data.

Geospatial Analysis of Safe Delivery App Events Based on Geographically Weighted Regression Tool

Abstract. This study explores the spatial relationship between the number of recorded events of the Safe Delivery App, which is a mobile learning tool to train midwives in developing countries, during three months, and several independent variables, including the number of health facilities, pregnancies, number of women of childbearing age, number of infants aged 0–1 years, mobile network coverage data and total population density. The study aims to identify and analyse the reach of the Safe Delivery App at the district level in Ghana country, considering the correlation between dependent and independent variables. The geospatial analysis of App usage events layered with several related explanatory variables is based on Geographically Weighted Regression. The explanatory variables were able to predict and explain the number of events with of accuracy score of 90 % at the district level. The results have provided valuable insights into the further roll-out of the App and helped to highlight the districts that need further support to roll out the Safe Delivery App considering the analysed independent variables.

A Usage-Based Proposal for Argument Structure of Directional Verbs in American Sign Language.

Verb agreement in signed languages has received substantial attention for a long time. Despite the numerous analyses about the linguistic status of verb agreement, there is little discussion about the argument structure associated with “directional verbs,” also known as agreeing/agreement or indicating verbs. This paper proposes a usage-based approach for analyzing argument structure constructions of directional verbs in American Sign Language (ASL). The proposal offers low-level constructions for reported speech, non-dedicated passive and reflexive, and stance verb constructions, which capture the patterns, abstracted from recurring usage events, that are part of users’ linguistic knowledge. The approach has potential to push the field of sign linguistics in new directions of understanding the interplay of language use and structure.

Insertion of the voiceless fricative [s] at word-final position in Japanese English

ABSTRACT This article explains how the voiceless fricative [s], or what we refer to as ‘special [s],’ occurs in free talk produced by Japanese speakers of English and what this means for the current understanding of Japanese English. The discussion is based on a small corpus compiled by the authors from recordings. The findings are threefold. The special [s] is always voiceless, occurs only word-finally, and arises from four different pragmatic forces. These forces are discourse marker, hesitation, cause and effect, and expression of self. The definition of Japanese English has been built heavily on the influence of the first language (L1) on the second language (L2). While L1 transfer is integral to L2 English production, the present article supplements this view by disclosing facts about what L2 users do while they are engaged in speaking in English; that is, in usage events.

A246 GASTROINTESTINAL ADVERSE EVENTS OF CANNABINOIDS IN PEDIATRIC POPULATIONS: A SYSTEMATIC REVIEW AND META-ANALYSIS

BackgroundUse of therapeutic cannabinoids in the pediatric population has increased in recent years within areas of epilepsy, palliative care, and cancer. Given the endocannabinoid system plays a crucial role in gastrointestinal (GI) homeostasis, exogenous cannabinoids have the potential to develop GI related side effects. While some GI adverse events have been reported with the use of therapeutic cannabinoids, the full profile of GI adverse events in the pediatric population is still unknown.AimsTo understand the impact of therapeutic cannabinoids on the GI system of pediatric patients, we performed a systematic review and meta-analysis to assess the prevalence of various GI-related adverse events arising from cannabinoid usage within pediatric populations.MethodsSearches were conducted from OVID MEDLINE, EMBASE, CINAHL, Web of Science, and The Cochrane Library for study screening. The included studies were quantitatively assessed for GI adverse events including nausea, diarrhea, increased appetite, decreased appetite, weight gain, weight loss, constipation, abdominal pain, and unspecified events. The prevalence along with their relation towards diagnosis, type of cannabinoid, dosage, duration of treatment, and study type were also assessed.ResultsAmong a total of 1201 patients across 25 included studies, an overall prevalence of 33.91% GI adverse events was observed in patients. The statistical analysis of the pediatric population displayed no relation between therapeutic cannabinoid usage and GI adverse events, bearing a large heterogeneity (I2 = 91%, P < 0.01). Upon analysis based on study type, a significant difference was observed between prospective and retrospective studies in categories of all GI adverse events, diarrhea, and decreased appetite, with prospective studies displaying a higher prevalence in all three categories. Nausea symptoms showed a significant negative correlation with maximum dosage of cannabinoid used. In addition, diarrhea was found to have a higher prevalence in Dravet syndrome related to an ion channel gene mutation (SCN1A) compared to other epilepsy patients.ConclusionsWhile the meta-analysis did not identify any significant associations between therapeutic cannabinoid usage and GI adverse events, there remains the potential for interactions of cannabinoids on GI symptoms depending on dosage and underlying patient factors.Funding AgenciesFarncombe Family Digestive Health Research Institute

WhatsNextApp: LSTM-Based Next-App Prediction With App Usage Sequences

Next app prediction can help enhance user interface design, pre-loading of apps, and network optimizations. Prior work has explored this topic, utilizing multiple different approaches but challenges like the user cold-start problem, data sparsity, and privacy concerns related to contextual data like location histories, persist. The user cold-start problem occurs when a user has recently registered to the smartphone app system and there is not enough information about his/her preferences and his/her history of smartphone usage. In this work, we try to address the above issues. We introduce WhatsNextApp, an approach based on LSTM (Long Short-Term Memory) networks using sequences of app usage logs. Our approach is inspired by Word Embeddings and treats sequences of app usage logs as sequences of words. We collect a real-life data set consisting of 975 Android users with over 22 million app usage events. We build a generic (user-independent) WhatsNextApp model and the evaluation with our data set shows that it outperforms related studies for existing users where we achieve a recall@8 (recall for the top 8 apps) of 92%. For the user cold-start problem with the 500 most frequent apps, we achieve a recall@8 of 82.7%.

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

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.