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

Analysis of Healthcare Seeking Behavior
Chrissy Dangel*1, Steven Allgeier1, Adam Haas2 and Amanda Johnson2

1US EPA, Cincinnati, OH, USA; 2CSC Science & Engineering, Cincinnati, OH, USA

Objective
This paper describes analyses of health seeking behaviors from 

two surveillance datastreams: Poison Control Center (PCC) calls 
and Emergency Department (ED) visit records. These analyses were 
conducted in order to quantify behaviors following the development 
of symptoms after water contamination exposure and to understand 
the motivation, decision-making and timing behind healthcare seeking 
behaviors.

Introduction
Healthcare seeking behavior is important to understand when 

interpreting public health surveillance data, planning for healthcare 
utilization, or attempting to estimate or model consequences of an 
adverse event, such as widespread water contamination. Although 
there is evidence that factors such as perceived susceptibility and 
benefits affect healthcare seeking behavior, it is difficult to develop 
accurate assumptions due to a lack of published research on this topic. 
Current conceptual behavior models, such as the health belief model, 
are not easily translated into quantifiable terms.

Methods
For the PCC analysis, data was included from 480 calls from two 

PCCs from September 2012 to April 2013. Shortly following the 
calls, PCC callers were surveyed and results were merged with call 
data. A subset of callers surveyed was preferentially selected based on 
the reason for call or substance described in an attempt to gather more 
information on situations similar to a possible water contamination 
event. The primary outcome of interest was the method of obtaining 
the PCC phone number. Methods of obtaining the PCC phone number 
were categorized as research, professional referral, PCC sticker/
magnet, friend/family and other/non-specified.

The ED visit records analysis comprised over 71,000 visit records 
from five EDs from January 2001 to December 2011 in Cincinnati, 
OH. In this analysis, the primary outcome of interest was means of 
arrival to the ED. ED arrival was dichotomized as Emergency Medical 
Services (EMS) utilization (yes/no). Healthcare seeking behaviors 
were categorized by characteristics such as age, disease category, 
insurance type, disposition as a severity indicator and time period.

For both data sets, exploratory data analysis was performed 
using contingency tables. Differences in the proportions of methods 
to obtain the PCC phone number and to means of ED arrival were 
compared across demographic variables and visit characteristics using 
chi-square tests of significance.

Results
The main finding from the PCC analysis was that insurance 

type was significantly associated with obtaining the PCC number 
(p =0.002). Survey respondents with private or employer based 
insurance were 1.9 times as likely to obtain the number from a PCC 
sticker or magnet while those with government based insurance were 
1.7 times as likely to obtain the number after a referral compared to 
other methods.

The ED study found that arrival to the ED by EMS was 
uncommon (<30%) but varied by ED (p <0.001). EMS utilization 
was more common in the autumn season and peaked on Fridays and 
Saturdays. Age, disease category, insurance type and disposition 

were significantly associated with EMS utilization (p’s <0.001). 
Adults at least 64 years of age were 1.9 times as likely to utilize EMS 
compared to younger adults. Children ages 15-19 were 2.8 times 
as likely to arrive by EMS compared to younger children. Patients 
arriving to the ED due to poisoning of drugs or toxic effects of non-
medical substances had the greatest EMS utilization. The effect of 
insurance type on EMS utilization varied by patient age. Adults with 
government based insurance and children with commercial based 
insurance were the most likely to utilize EMS. Disposition exhibited 
the strongest effect on EMS utilization. Children who were admitted 
were 7.7 times as likely to arrive by EMS. Adult patients were 5.7 
times as likely to arrive by EMS if they were admitted. In a sensitivity 
analysis adjusting for disposition, the association between insurance 
type and EMS utilization was no longer significant for admitted 
patients (p =0.105).

Conclusions
This study better defines the health seeking behaviors of patients 

experiencing symptoms analogous to those following water 
contamination and highlights the importance of using diverse outreach 
methods to promote PCC and EMS utilization.

Keywords
healthcare; behavior; poison control center; emergency department; 
analysis

References

1. Sheeren, P. & Abraham, C The Health Belief Model, in predicting
Health Behaviour (Conner, M. & Norman, P. eds). Buckingham: Open 
University Press. 1995

*Chrissy Dangel
E-mail: dangel.chrissy@epa.gov    

Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org * 7(1):e17, 2015