Acute Coronary Syndrome and Care-Seeking Delay: A Web Based Behavioral Study

Overview

The primary aim of this study is to increase our understanding of care-seeking behavior surrounding heart attacks or acute coronary syndromes \[ACS\]. This study uses an internet based survey to ask individuals how they obtained medical care in the midst of a heart attack. At present, care-seeking delay among individuals stricken with a heart attack prevents them from obtaining the full therapeutic benefit of hospital based medical care in a timely manner to reduce the long term health consequences of a heart attack. By using a self-tailoring survey instrument the study attempts to take into consideration the complex social processes by which the individual and their family make decisions to seek medical care for symptoms of a heart attack. The study is designed to obtain a national sample of ACS care-seeking behavior in the United States.

The primary aim of this study is to increase our understanding of care-seeking behavior surrounding acute coronary syndromes \[ACS\] through the application and testing of an integrated self-regulatory care-seeking model \[ISCM\] and the utilization of a world wide web \[WWW\] based ISCM survey instrument. At present, care-seeking delay among individuals stricken with ACS \[acute myocardial infarction \[AMI\], unstable angina or sudden cardiac death\] prevents many patients from obtaining the full therapeutic benefit of hospital based thrombolytic and/or mechanical reperfusion therapy to reduce the morbid and mortal consequences of acute cardiovascular disease \[CVD\]. Results from prior studies of ACS care-seeking delay generally rely on measures of time duration from acute symptom onset \[ASO\] to hospital emergency department \[ED\] arrival and measures of social demographic and clinical variables, extracted from emergency medical system \[EMS\] logs and/or ED charts. By relying on such limited data sources, these studies have failed to take into consideration the complexity of the social and behavioral processes by which individuals make decisions to seek medical care for symptoms of ACS. By applying and testing an integrated self-regulatory care-seeking model \[ISCM\] to ACS care-seeking, it will be possible to delineate decision points and situations and circumstances that are critical to producing patterns of care-seeking that are efficient and expeditious or are protracted and delayed. The care-seeking process in the ISCM is viewed in terms of five analytic care-seeking phases from warning or premonitory symptoms to ED arrival. By noting the presence or absence of three central phases, it is possible to distinguish patterns of care-seeking and to record the duration of each care-seeking phase and total duration of each care-seeking path to the ED. The ISCM assumes that coping behaviors emerge over time and initially care-seeking behaviors are guided by demographic and structural factors. As the self-regulatory and coping processes emerge over the course of the ACS episode, the influence of these factors diminishes and emergent ACS evaluations, advice from others and health care providers, emerging symptoms and emotions, come to dominate the care-seeking process and directly determine the duration of ACS care-seeking. The study will test and explore the following hypotheses: 1. Commonly used demographic variables, such as, age, sex, education and race will prove to be only one component, along with social situational and circumstantial, cognitive and emotional variables, that significantly influence the course and duration of ACS care-seeking. 2. The ISCM will show that decision process variables associated with the care-seeking phases, the Self, Lay and Health Care Provider Phases, as shown in the Figure, are important predictors of time from symptoms onset to ED arrival and in the prediction of the length of time spent in each care-seeking phase. 3. The ISCM will show differences in the total duration of the 5 most frequently occurring care-seeking paths to the ED and will explain most of the differences by the variables in Hypotheses 1 and 2. 4. Conditional models of subjects who use key care-seeking phases, namely the Self, Lay and Health Care Provider Phases, will identify unique predictors of time in each phase. 5. Significant interaction effects \[model coefficient differences\] will be found for important risk variables and socially important gender, age and racial groups. For example, African-Americans will show differences in model coefficients from Non-African-Americans. The proposed study utilizes a twice tested pilot ISCM survey instrument placed on two secure dedicated web servers to collect patient care-seeking data. The target sample size for this study is 2,314 patients who have experienced an ACS episode. Subjects will be recruited using: 1\] extensive marketing of the study URL on websites of interest to cardiac patients and oriented toward cardiac health and ACS risks, such as, diabetes or hypertension; 2\] placement of study information on discussion boards and chat rooms with a cardiac health focus; 3\] emailing of a study flyer PDF for posting in public libraries, senior citizen centers, African-American churches, county and municipal health clinics, and HMO web sites; 4\] application of a viral marketing strategy; and 5\] use of selected print media, for example, AARP publications. Marketing of the study site URL will emphasize the recruitment of women, the elderly and minorities. In the data collection phase of the study, the proportions of myocardial infarction and angina pectoris subjects falling into each cell of the cross classification of age, sex and race will be derived from NHANES 1999-2000 and 2001-2002 studies. These estimates of the population proportions in these cells will be used to determine quotas for each of the cells in an attempt to ensure all risk groups are adequately represented. We will over sample African-Americans. Statistical models will be used to estimate the effects of variables that characterize the decision process on the time delay until ED arrival. Time required to move from one phase of the decision process to the next will also be modeled, as will the different paths chosen by the subjects. Models will also include patient characteristics and other circumstantial and situational variables in the decision process.