Social Interactions and Activities of Latino ED Users

Columbia University Health Sciences, New York NY

Investigators

Abstract

DESCRIPTION (provided by the applicant): Social interactions are significant to a myriad of phenomena, since many are dependent upon the development and maintenance of personal or social ties. With regard to help-seeking and health care utilization patterns during an illness episode, patients often consult with family, friends, community members, and others for advice (McKinlay, 1973; Pilisuk and Froland, 1978; Salloway and Dillon, 1973). Models of illness and help-seeking behavior attempt to explain how individuals recognize and respond to illnesses, but have struggled to find a balance between factors (indirect and direct) that are related to health care use and the underlying social processes that influence illness behaviors. The use of alternative methods of care or social interactions, which are equally important in understanding how and why individuals respond to an illness, are often not incorporated in dominant models of utilization. Increasingly, attention has been given to a dynamic framework that posits that help-seeking and decision-making processes are embedded in social interactions and activities among relevant actors (Uehara, 2001; Pescosolido, 1992). A recently developed model by Pescosolido, the Network-Episode Model (NEM), astutely states that health problems are part of a social process that is recognized, managed, and acted upon through contacts that the patient has in the community, within health care settings, or social service agencies (Pescosolido and Boyer, 1999). This study applies the NEM model to examine the help-seeking behaviors of Latino adult emergency department users in New York City. Specifically, this dissertation will use a dynamic, help-seeking model to identify the social strategies or patterns of help-seeking that Latinos undertake immediately prior to an ED visit and examine the correlates of these strategies or patterns. In addition, the effect of the strategies or patterns and other selected variables on avoidable or unavoidable medical conditions among Latinos will be explored. This study is a secondary analysis of survey data from individuals who have visited one of four participating emergency departments (EDs) in New York City. In order to predict the determinants of the strategies or patterns ED patients use prior to going to the ED, a multinomial Iogit (MNL) will be used for analysis due to the complex nature of the dependent variable (i.e., multiple categories). In addition, logistic regression will be utilized to examine the predictors of avoidable or unavoidable ED use for Latino patients in this study. Identifying and investigating the underlying factors involved in ED use patterns will assist in the development and implementation of effective policy initiatives and interventions for many vulnerable populations in urban settings. Examining this issue in New York City provides a unique opportunity, as it has one of the largest and most diverse immigrant populations in the country and an extensive health care delivery system.