Towards Expectation Detection in Language: A Case Study on Treatment Expectations in Reddit

TL;DR

This study introduces Expectation Detection in natural language, creating a Reddit-based corpus to analyze patient treatment expectations, revealing linguistic patterns and differences across health contexts.

Contribution

It presents the first NLP-focused task for detecting expectations in online health discussions and introduces the RedHOTExpect corpus for future research.

Findings

Expectations are more optimistic in physical health posts.

Patients mainly discuss benefits over negative outcomes.

Linguistic patterns distinguish expectation types.

Abstract

Patients' expectations towards their treatment have a substantial effect on the treatments' success. While primarily studied in clinical settings, online patient platforms like medical subreddits may hold complementary insights: treatment expectations that patients feel unnecessary or uncomfortable to share elsewhere. Despite this, no studies examine what type of expectations users discuss online and how they express them. Presumably this is because expectations have not been studied in natural language processing (NLP) before. Therefore, we introduce the task of Expectation Detection, arguing that expectations are relevant for many applications, including opinion mining and product design. Subsequently, we present a case study for the medical domain, where expectations are particularly crucial to extract. We contribute RedHOTExpect, a corpus of Reddit posts (4.5K posts) to study expectations in this context. We use a large language model (LLM) to silver-label the data and validate its quality manually (label accuracy ~78%). Based on this, we analyze which linguistic patterns characterize expectations and explore what patients expect and why. We find that optimism and proactive framing are more pronounced in posts about physical or treatment-related illnesses compared to mental-health contexts, and that in our dataset, patients mostly discuss benefits rather than negative outcomes. The RedHOTExpect corpus can be obtained from https://www.ims.uni-stuttgart.de/data/RedHOTExpect