Identifying social isolation themes in NVDRS text narratives using topic modeling and text-classification methods

2025-06-18

AI Summary

This study addresses the public health challenge of monitoring social isolation—particularly hindered by missing structured fields in the National Violent Death Reporting System (NVDRS). We propose a two-stage NLP framework: first, domain-specific lexicon construction via LDA topic modeling; second, hybrid classification integrating fine-tuned BERT with interpretable logistic regression. Applied to over 300,000 suicide narrative texts, it is the first systematic identification and quantification of chronic versus situational social isolation. The model achieves an F1-score of 0.86 and accuracy of 0.82, identifying 1,198 chronic isolation cases. Significant associations are found with gender, sexual orientation, and marital status. Our approach overcomes the bottleneck of extracting latent social determinants from unstructured death investigation narratives, offering a scalable, interpretable technical pathway for proactive surveillance and precision intervention targeting loneliness-related suicide risk.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Application Domains: Humanities & Computational Social Science
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

Social isolation and loneliness, which have been increasing in recent years strongly contribute toward suicide rates. Although social isolation and loneliness are not currently recorded within the US National Violent Death Reporting System's (NVDRS) structured variables, natural language processing (NLP) techniques can be used to identify these constructs in law enforcement and coroner medical examiner narratives. Using topic modeling to generate lexicon development and supervised learning classifiers, we developed high-quality classifiers (average F1: .86, accuracy: .82). Evaluating over 300,000 suicides from 2002 to 2020, we identified 1,198 mentioning chronic social isolation. Decedents had higher odds of chronic social isolation classification if they were men (OR = 1.44; CI: 1.24, 1.69, p<.0001), gay (OR = 3.68; 1.97, 6.33, p<.0001), or were divorced (OR = 3.34; 2.68, 4.19, p<.0001). We found significant predictors for other social isolation topics of recent or impending divorce, child custody loss, eviction or recent move, and break-up. Our methods can improve surveillance and prevention of social isolation and loneliness in the United States.

Problem

• Detect social isolation in NVDRS narratives using NLP
• Develop classifiers to identify chronic social isolation cases
• Analyze predictors like gender and divorce for isolation risks

Innovation

• Using topic modeling for lexicon development
• Applying supervised learning classifiers for identification
• Analyzing NVDRS narratives with NLP techniques