Use of Near-Real–Time Data to Inform Underage Drinking Surveillance in Nebraska

Sandra Gonzlez, David DeVries, Ming Qu

Abstract


ObjectiveThe objective of this pilot study was to develop and evaluatesyndromic definitions for the monitoring of alcohol-related emergencydepartment (ED) visits in near-real–time syndromic surveillance(SyS) data. This study also evaluates the utility of SyS ED data forthe monitoring of underage drinking.IntroductionUnderage drinking is a significant public health problem in theUnited States as well as in Nebraska1-2. Alcohol consumption amongunderage youth accounts for approximately 5,000 deaths each yearin the United States, including motor vehicle crash related deaths,homicides and suicides1. In Nebraska, 23% of 12-20 year olds havereported alcohol use during the past 30 days3. In 2010, the estimatedtotal costs of underage drinking in Nebraska were $423 million.These costs included medical care, work loss along with pain andsuffering2. The health consequences of underage drinking includealcohol-related motor vehicle crashes and other unintentional injuries,physical and sexual assault, suicide, self-inflicted injury, death fromalcohol poisoning, and abuse of other drugs1, 4. The monitoring ofnear-real–time ED data could help underage drinking preventionefforts by providing timelier actionable public health information.MethodsNebraska SyS data from 32 ED facilities was analyzed for visitsof 12 to 20 year olds during October 1, 2015 to August 31, 2016.Three syndromic definitions were developed and tested for themonitoring of alcohol-related ED visits in near-real–time SyS databy using ESSENCE. The first and second definitions were based onquerying the chief complaint (CC) field for search terms associatedwith alcohol use and alcohol abuse or intoxication respectively.The third definition consisted of ICD-9-CM and ICD-10-CMdiagnostic codes associated to alcohol abuse or intoxication. Thesethree definitions were evaluated for internal consistency: reporteddiagnostic codes were used to evaluate the first and second definition,while text in the CC field was used to evaluate the third definition.Records with missing CC or diagnostic codes were excluded from theconsistency analysis. In addition, the CC field of records detected bythe third definition was evaluated for possible alcohol-related healthconsequences.ResultsA total of 126 cases were detected by using the first definition(CC search terms for alcohol use); 61% (50/82) of these identifiedalcohol abuse-related diagnostic codes. On the other hand, a totalof 64 cases were detected by using second definition (CC searchterms specific for alcohol abuse or intoxication); 89% (33/37) ofthese identified alcohol abuse-related diagnostic codes. The thirddefinition (diagnostic codes only) detected 111 cases; 49% (51/105)of these identified alcohol-related search terms in records withreported CC. However, keywords associated to alcohol-related healthconsequences, such as injury, assault, and use of other drugs werefound in records with no alcohol-related search terms in the CC field.Diagnostic codes associated to alcohol-related health consequenceswere observed in 93% (50/54) of these records. These results indicatethat alcohol use is underreported in the CC field.ConclusionsA higher internal consistency was observed for the syndromicdefinition based on CC search terms associated with alcohol abuseor intoxication. However, a syndromic definition based on diagnosticcodes is preferred due to the underreporting of alcohol use in the CCfield. The detection of underage alcohol use-related cases could beimproved by adding alcohol abuse or intoxication CC search termsto a syndromic definition based on diagnostic codes. Overall, resultsof this pilot study suggest that a syndromic definition based ondiagnostic codes can potentially enhance the surveillance of underagedrinking and alcohol-related health consequences.

Full Text:

PDF


DOI: https://doi.org/10.5210/ojphi.v9i1.7732



Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org