An Unsupervised Natural Language Processing Pipeline for Assessing Referral Appropriateness

TL;DR

This paper introduces an unsupervised NLP pipeline that accurately assesses the appropriateness of medical referrals from free text, aiding healthcare policy and reducing unnecessary procedures.

Contribution

It presents a novel unsupervised NLP method using Transformer embeddings to evaluate referral appropriateness without labeled data, applicable across different examination types.

Findings

High precision and recall in identifying referral reasons.

Effective detection of inappropriate referrals and regional variation.

Supported policy changes in Lombardy healthcare system.

Abstract

Objective: Assessing the appropriateness of diagnostic referrals is critical for improving healthcare efficiency and reducing unnecessary procedures. However, this task becomes challenging when referral reasons are recorded only as free text rather than structured codes, like in the Italian NHS. To address this gap, we propose a fully unsupervised Natural Language Processing (NLP) pipeline capable of extracting and evaluating referral reasons without relying on labelled datasets. Methods: Our pipeline leverages Transformer-based embeddings pre-trained on Italian medical texts to cluster referral reasons and assess their alignment with appropriateness guidelines. It operates in an unsupervised setting and is designed to generalize across different examination types. We analyzed two complete regional datasets from the Lombardy Region (Italy), covering all referrals between 2019 and 2021 for venous echocolordoppler of the lower limbs (ECD;n=496,971; development) and flexible endoscope colonoscopy (FEC; n=407,949; testing only). For both, a random sample of 1,000 referrals was manually annotated to measure performance. Results: The pipeline achieved high performance in identifying referral reasons (Prec=92.43% (ECD), 93.59% (FEC); Rec=83.28% (ECD), 92.70% (FEC)) and appropriateness (Prec=93.58% (ECD), 94.66% (FEC); Rec=91.52% (ECD), 93.96% (FEC)). At the regional level, the analysis identified relevant inappropriate referral groups and variation across contexts, findings that informed a new Lombardy Region resolution to reinforce guideline adherence. Conclusions: This study presents a robust, scalable, unsupervised NLP pipeline for assessing referral appropriateness in large, real-world datasets. It demonstrates how such data can be effectively leveraged, providing public health authorities with a deployable AI tool to monitor practices and support evidence-based policy.