|
Chiung M. Chen, MA, Hsiao-ye Yi, PhD, and Thomas C. Harford. Alcohol Epidemiologic Data System, NIH/NIAAA, CSR, Incorporated, 2107 Wilson Boulevard, Suite 1000, Arlington, VA 22201, 7037417125, cchen@csrincorporated.com
Previous reviews of the epidemiologic evidence adopt a contextual paradigm in which types of injury may be grouped under four ecological categories: residential, recreational, occupational, and transportation-related injuries (Smith and Kraus, 1988). The current study examines the extent to which drinking relates to these injury categories and how the relationship is modified by age.
4,786 injury episodes reported by 4,549 adult respondents were selected from the 1997–2001 National Health Interview Survey. The episodes were original coded by 9 external causes of injury (e.g., fall, fire), 18 locations (e.g., school, street), and 11 activities (e.g., driving, cooking). Based on these codings, each episode was classified into one of the four ecological categories according to its highest posterior probability, using latent class analysis. Multinomial logistic regression was subsequently used to relate the four latent classes to two alcohol consumption measures—(1) drinking level based on average daily intake, and (2) episodic heavy drinking based on the frequency of having 5+ drinks per sitting in the past year—over age, controlling for sex, race, employment, and marital status.
Preliminary results indicated that, compared with non-drinkers, the odds favoring occupational and residential injuries over transportation-related injuries were generally higher for drinkers at different levels (although not always significant) as well as for episodic heavy drinkers. The odds favoring recreational injuries over transportation-related injuries increased with drinking levels, but decreased with episodic heavy drinking. Additional models tested interactions between drinking measures and age to examine how these relationships vary over the life cycle.
Learning Objectives:
Keywords: Alcohol Use, Injuries
Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.