Using n-Grams for Syndromic Surveillance in a Turkish Emergency Department Without English Translation: A Feasibility Study.

INTRODUCTION: Syndromic surveillance is designed for early detection of disease outbreaks. An important data source for syndromic surveillance is free-text chief complaints (CCs), which are generally recorded in the local language. For automated syndromic surveillance, CCs must be classified into predefined syndromic categories. The n-gram classifier is created by using text fragments to measure associations between chief complaints (CC) and a syndromic grouping of ICD codes. OBJECTIVES: The objective was to create a Turkish n-gram CC classifier for the respiratory syndrome and then compare daily volumes between the n-gram CC classifier and a respiratory ICD-10 code grouping on a test set of data. METHODS: The design was a feasibility study based on retrospective cohort data. The setting was a university hospital emergency department (ED) in Turkey. Included were all ED visits in the 2002 database of this hospital. Two of the authors created a respiratory grouping of International Classification of Diseases, 10th Revision ICD-10-CM codes by consensus, chosen to be similar to a standard respiratory (RESP) grouping of ICD codes created by the Electronic Surveillance System for Early Notification of Community-based Epidemics (ESSENCE), a project of the Centers for Disease Control and Prevention. An n-gram method adapted from AT&T Labs' technologies was applied to the first 10 months of data as a training set to create a Turkish CC RESP classifier. The classifier was then tested on the subsequent 2 months of visits to generate a time series graph and determine the correlation with daily volumes measured by the CC classifier versus the RESP ICD-10 grouping. RESULTS: The Turkish ED database contained 30,157 visits. The correlation (R (2)) of n-gram versus ICD-10 for the test set was 0.78. CONCLUSION: The n-gram method automatically created a CC RESP classifier of the Turkish CCs that performed similarly to the ICD-10 RESP grouping. The n-gram technique has the advantage of systematic, consistent, and rapid deployment as well as language independence.

The ngram chief complaint classifier: a novel method of automatically creating chief complaint classifiers based on international classification of diseases groupings.

INTRODUCTION: The ngram classifier is created by using text fragments to measure associations between chief complaints (CC) and a syndromic grouping of ICD-9-CM codes. OBJECTIVES: For gastrointestinal (GI) syndrome to determine: (1) ngram CC classifier sensitivity/specificity. (2) Daily volumes for ngram CC and ICD-9-CM classifiers. DESIGN: Retrospective cohort. SETTING: 19 Emergency Departments. PARTICIPANTS: Consecutive visits (1/1/2000-12/31/2005). PROTOCOL: (1) Used an existing ICD-9-CM filter for "lower GI" to create the ngram CC classifier from a training set and then measured sensitivity/specificity in a test set using an ICD-9-CM classifier as criterion. (2) Compare daily volumes based on ICD-9-CM with that predicted by the ngram classifier. RESULTS: For a specificity of 0.96, sensitivity was 0.70. The daily volume correlation for ngram vs. ICD-9-CM was R=0.92. CONCLUSION: The ngram CC classifier performed similarly to manually developed CC classifiers and has advantages of rapid automated creation and updating, and may be used independent of language or dialect.

Data mining a diabetic data warehouse.

Diabetes is a major health problem in the United States. There is a long history of diabetic registries and databases with systematically collected patient information. We examine one such diabetic data warehouse, showing a method of applying data mining techniques, and some of the data issues, analysis problems, and results. The diabetic data warehouse is from a large integrated health care system in the New Orleans area with 30,383 diabetic patients. Methods for translating a complex relational database with time series and sequencing information to a flat file suitable for data mining are challenging. We discuss two variables in detail, a comorbidity index and the HgbA1c, a measure of glycemic control related to outcomes. We used the classification tree approach in Classification and Regression Trees (CART) with a binary target variable of HgbA1c >9.5 and 10 predictors: age, sex, emergency department visits, office visits, comorbidity index, dyslipidemia, hypertension, cardiovascular disease, retinopathy, end-stage renal disease. Unexpectedly, the most important variable associated with bad glycemic control is younger age, not the comorbiditity index or whether patients have related diseases. If we want to target diabetics with bad HgbA1c values, the odds of finding them is 3.2 times as high in those <65 years of age than those older. Data mining can discover novel associations that are useful to clinicians and administrators [corrected].