Google Trends Can Predict Which Patients Will Visit EDs
Abstract
Graph 1.: Google, trends, ED, emergency care, AHRQ, algorithms, machine learning, LASSO, coefficientsFigureHave you ever wanted to peek into our patients' search histories to see which questions they had for Dr. Google? Google first made search queries public in 2006 via a tool called Google Trends. The tool allows users to type in a phrase to see the relative popularity of this term over a timespan. Researchers have attempted to use it to predict everything from the performance of the stock market to the spread of influenza, often with some success. Let's look at a search of particular interest to us: “emergency room near me,” a query that strongly suggests an active search for emergency care. This search had a few spikes during the COVID peaks, and it was associated with some clear seasonality. (Graph 1.) But how can we use this to see which other searches users are querying when they search for an ED. Creating a Model The first step in the process was data acquisition. Google does not allow queries for daily data from Trends over long time spans, and it does not provide absolute search counts, only scaled search counts over a period. I had to link multiple queries using overlapping periods to produce data that would be scaled similarly over the whole span to acquire long-term data by symptom. I composed a list of 34 symptoms based on the AHRQ ED visit data from 2018, a difficult task because these needed to be rephrased as search terms (a subjective process). The second…
Read the paper
Related studies
- The Matrisome: In Silico Definition and In Vivo Characterization by Proteomics of Normal and Tumor Extracellular Matrices
- The Escherichia coli MG1655 in silico metabolic genotype: Its definition, characteristics, and capabilities
- In silico pharmacology for drug discovery: methods for virtual ligand screening and profiling
- In silico prediction of splice-altering single nucleotide variants in the human genome
- In silico clinical trials: how computer simulation will transform the biomedical industry
- In silico optimization of a guava antimicrobial peptide enables combinatorial exploration for peptide design
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.