Physician-level determinants of HCV screening during pregnancy in a U.S. sample

Optimization of hepatological clinical guidelines interpretation by large language models: a retrieval augmented generation-based framework

Large language models (LLMs) can potentially transform healthcare, particularly in providing the right information to the right provider at the right time in the hospital workflow. This study investigates the integration of LLMs into healthcare, specifically focusing on improving clinical decision support systems (CDSSs) through accurate interpretation of medical guidelines for chronic Hepatitis C Virus infection management. Utilizing OpenAI's GPT-4 Turbo model, we developed a customized LLM framework that incorporates retrieval augmented generation (RAG) and prompt engineering. Our framework involved guideline conversion into the best-structured format that can be efficiently processed by LLMs to provide the most accurate output. An ablation study was conducted to evaluate the impact of different formatting and learning strategies on the LLM's answer generation accuracy. The baseline GPT-4 Turbo model's performance was compared against five experimental setups with increasing levels of complexity: inclusion of in-context guidelines, guideline reformatting, and implementation of few-shot learning. Our primary outcome was the qualitative assessment of accuracy based on expert review, while secondary outcomes included the quantitative measurement of similarity of LLM-generated responses to expert-provided answers using text-similarity scores. The results showed a significant improvement in accuracy from 43 to 99% (p < 0.001), when guidelines were provided as context in a coherent corpus of text and non-text sources were converted into text. In addition, few-shot learning did not seem to improve overall accuracy. The study highlights that structured guideline reformatting and advanced prompt engineering (data quality vs. data quantity) can enhance the efficacy of LLM integrations to CDSSs for guideline delivery.

Maternal neighbourhood-level social determinants of health and their association with paediatric hepatitis C screening among children exposed to hepatitis C in pregnancy

BACKGROUND

Current guidelines recommend HCV screening by 18 months of age for those exposed to HCV in utero; yet, screening occurs in the minority of children.

OBJECTIVES

To evaluate the association between maternal neighbourhood-level social determinants of health (SDOH) and paediatric HCV screening in the general population in a publicly funded healthcare system in Canada.

METHODS

Retrospective cohort study using administrative healthcare data held at ICES. Children born to individuals positive for HCV RNA in pregnancy from 2000 to 2016 were identified and followed for 2 years. Major SDOH were identified, and the primary outcome was HCV screening in exposed children (HCV antibody and/or RNA). Associations between SDOH and HCV screening were determined using multivariate Poisson regression models adjusting for confounding.

RESULTS

A total of 1780 children born to persons with +HCV RNA were identified, and 29% (n = 516) were screened for HCV by age two. Most mothers resided in the lowest income quintile (42%), and most vulnerable quintiles for material deprivation (41%), housing instability (38%) and ethnic diversity (26%) with 11% living in rural locations. After adjustment for confounding, maternal rural residence (risk ratio [RR] 0.82, 95% confidence interval [CI] 0.62, 1.07) and living in the highest dependency quintile (RR 0.83, 95% CI 0.65, 1.07) were the SDOH most associated with paediatric HCV screening. Younger maternal age (RR 0.98 per 1-year increase, 95% CI 0.97, 0.99), HIV co-infection (RR 1.69, 95% CI 1.16, 2.48) and GI specialist involvement (RR 1.18, 95% CI 1.00, 1.39) were associated with higher probabilities of screening.

CONCLUSIONS

Among children exposed to HCV during pregnancy, rural residences and living in highly dependent neighbourhoods showed a potential association with a lower probability of HCV screening by the age of 2. Future work evaluating barriers to paediatric HCV screening among rural residing and dependent residents is needed to enhance the screening.