Venous thromboembolism (VTE) contributes to hospitalisation-associated morbidity. Although guidelines recommend limiting VTE prophylaxis to high-risk patients, some physicians prescribe it broadly. We compared beliefs of low and high prescribing physicians.

We surveyed hospitalists and medical residents who had the opportunity to prescribe prophylaxis ≥50 times. Best-worst scaling was used to assess their beliefs. Using a balanced incomplete block design, we created seven choice tasks with seven statements regarding prophylaxis beliefs each presented four times. For each task, physicians selected the statement that most and least reflected their beliefs. We used a count method to calculate best-worst scores and a conditional logistic regression choice model to compare low and high prescribers.

Of 434 invitees, 172 (40%) completed all survey questions between June and November 2023. Low (n=86, ≤62.5% prescribing rate) and high (n=86, >62.5 prescribing rate) prescribers endorsed similar beliefs with differing levels of agreement. All felt confident to prescribe prophylaxis appropriately (low: +1.13, high: +1.10, p=0.81). High prescribers expressed more concern about VTE without prophylaxis (+1.02 vs +0.65, p=0.002). Low prescribers disagreed more that prophylaxis had no downside (–1.03 vs –0.73, p=0.01). High prescribers worried less about prophylaxis risks (–0.49 vs –0.22, p=0.01), and overuse (–0.61 vs –0.34, p=0.02).

Compared with low prescribers, high prescribers were more concerned about VTE without prophylaxis and less about harms. These differences in beliefs may underlie physician behaviour and could be targets for interventions to reduce inappropriate prophylaxis.

Exposure to prescription opioids following traumatic injury can increase the risk of developing tolerance, persistent opioid use and opioid use disorder. The mechanisms underlying opioid tolerance or dependence are not well understood, and no biomarkers predict risk. Opioid exposure causes epigenetic modifications, including alterations in microRNA (miRNA) expression. Several miRNAs, which regulate synaptic plasticity, are hypothesised to underlie substance use disorders and influence µ-opioid receptor levels, modulating opioid tolerance. This project aims to develop a bio-behavioural signature to predict persistent opioid use and chronic pain up to 6 months post-discharge.

The study will use a prospective cohort design, enrolling 180 adult patients at a Level I Trauma Center who are prescribed opioids at discharge. Prospective data will be collected in the hospital and at 7 days and 1, 3 and 6 months post-discharge. Biological data (genotyping and miRNA levels) and clinical measures of opioid use, pain, pain sensitivity (EEG) and psychosocial functioning will be collected at each time point. Bayesian regression methods will be used to identify baseline clinical, genetic, epigenetic and psychosocial predictors of opioid use and pain outcomes at 6 months post-discharge. Growth mixture modelling will identify distinct subgroups with varying trajectories, followed by Bayesian hierarchical modelling to predict trajectory classification based on predictor variables.

Ethics approval for this study was obtained from the University of Texas Health Science Center at Houston Committee for the Protection of Human Subjects (HSC-MS-24–0314). Findings will be disseminated in peer-reviewed scientific journals and at national and international conferences.