2014.ISDS.Abstracts.Final.pdf


ISDS Annual Conference Proceedings 2014. This is an Open Access article distributed under the terms of the Creative Commons Attribution-
Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution, 
and reproduction in any medium, provided the original work is properly cited.

ISDS 2014 Conference Abstracts

Using a Syndromic Surveillance System to Evaluate the 
Impact of a Change in Alcohol Law
Julia A. Dilley*1, Atar Baer2, Jeff Duchin2 and Julie E. Maher1

1Program Design & Evaluation Services, Multnomah County Health Department and Oregon Public Health Division, Portland, OR, 
USA; 2Public Health - Seattle & King County, Seattle, WA, USA

Objective
To determine whether there were changes in alcohol-related 

emergency department (ED) visits in Washington State associated 
with statewide alcohol system deregulation.

Introduction
In November 2011, Washington State voters passed Initiative 1183 

(I-1183) which closed state-owned and contracted liquor stores and 
opened the market for “hard liquor” sales in the private sector. The 
change in law was implemented on June 1, 2012. Increases in alcohol-
related ED visits were postulated as one potential impact if there was 
increased alcohol use or excessive consumption associated with the 
change in law.

Methods
We examined weekly counts of ED visits for alcohol-related 

conditions from all but one hospital in King County, Washington, 
for the period January 2010 – September 2013 (16 months after 
implementation of the law). King County includes approximately 
one-third of the state population and ED data are provided to the 
county as part of ongoing syndromic surveillance. Outcomes were 
ED visits with an alcohol-related chief complaint or diagnosis code. 
Chief complaint search strings included terms such as: alcohol, drunk, 
names of alcohol types (e.g., beer, whiskey, vodka), alcohol on 
breath. Diagnosis codes included ICD-9 codes previously described 
by the Centers for Disease Control and Prevention (CDC) for acute 
impacts of excessive alcohol consumption (see http://apps.nccd.cdc.
gov/DACH_ARDI/Info/ICDCodes.aspx). We used a linear regression 
model with a spline at June 1, 2012, to determine if implementation 
of I-1183 was associated with changes in weekly ED visits. We 
fitted a series of regression models, stratified by both age groups 
(<21, 21-39, 40+) and gender. The main covariates of interest were a 
binary “policy” coefficient with values of “0” prior to June 2012 and 
“1” including and after June 2012, and an interaction term (policy 
coefficient x week). The dependent variables in the regression 
models were log-transformed to achieve homogeneity in the error 
structure. The model also included covariates for background trend 
(time variable for week), seasonality (calendar quarters), and general 
healthcare access (total ED visits each week for any reason). Our 
models took into account autocorrelated error terms by using an 
autoregressive modeling approach with Newey-West standard errors 
and a lag of 12 in Stata v. 13.0.

Results
Statistically significant (p<.05) increases in alcohol-related ED 

visits post-law were observed among minors (age <21) and adults 
ages 40 and older. The initial increases among minors, however, 
appeared to decline over time. No significant changes in alcohol-
related ED visits were observed among adults ages 21-39. Findings 
may be limited by ecological bias as well as the choice of search 
terms included in the alcohol syndrome definition. Future validation 
of the syndrome definition may improve the sensitivity/specificity of 
the results. Both alcohol-related ED visits as a percentage of total 

visits and also counts of alcohol-related ED visits (with adjustment for 
total visits) were independently explored as the outcomes of interest. 
Patterns of significance were similar. Trend analysis must incorporate 
adjustment for known data quality issues in syndromic surveillance, 
such as data drop-offs or new hospitals being added during the 
evaluation period.

Conclusions
Alcohol-related ED visits increased among minors (age younger 

than 21) and adults ages 40 and older after changes in Washington 
State law increased availability of “hard liquor.” Syndromic 
surveillance data were useful for describing public health impact of 
a change in law.

Keywords
Alcohol; Emergency department data; Public Health Law

Acknowledgments

This study was supported by Public Health Law Research, a national 
program of the Robert Wood Johnson Foundation.

*Julia A. Dilley
E-mail: julia.dilley@state.or.us    

Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org * 7(1):e70, 2015