Racial/Ethnic and Educational Differences in the Estimated Odds of Recent Nitrite Use among Adult Household Residents in the United States: An Illustration of Matching and Conditional Logistic Regression

Abstract

This article compares estimates of the relative odds of nitrite use obtained from weighted unconditional logistic regression with estimates obtained from conditional logistic regression after post-stratification and matching of cases with controls by neighborhood of residence. We illustrate these methods by comparing the odds associated with nitrite use among adults of four racial/ethnic groups, with and without a high school education. We used aggregated data from the 1994-B through 1996 National Household Survey on Drug Abuse (NHSDA). Difference between the methods and implications for analysis and inference are discussed.

Family Attention and Tobacco Smoking Among Adolescents in Central America, Panama, and the Dominican Republic

“…As a result, all tobacco nonusers attending the same schools and living in the same areas as tobacco users serve as controls. This epidemiological approach to forming risk sets or matched strata defined by neighborhoods or area of sampling and recruitment, and the associated biostatistical methods, have been described in detail by senior biostatisticians and epidemiologists such as Breslow and Day (1980), and this approach has been used extensively in prior epidemiological studies of drug use to account for the potential distorting influences of unmeasured neighborhood characteristics (e.g., Delva et al, 2000;Lillie-Blanton et al, 1993).…”

Violent victimization and drug involvement among Mexican middle school students

“…Post‐stratification by home‐room and conditional logistic regression was used to remove potentially affecting differences across schools, grades and home‐rooms (e.g. neighborhood differences, teacher’s policies and management of violence behaviors, among others), as well as to adjust simultaneously for imbalances in age, sex and other covariates that might confound the association of drug involvement and violent victimization (Anthony, Tien & Petronis 1989; Lillie‐Blanton, Schuster & Anthony 1993; Wagner et al 1994; Delva, Spencer & Lin 2000; Wagner & Anthony 2002).…”

A community prevention model to prevent children from inhaling and ingesting harmful legal products

“…Moreover, few programs focusing on HLPs have been developed or evaluated, and few models are available to guide prevention efforts dealing with this substance abuse problem (Brouette & Anton, 2001; Delva, Spencer, & Lin, 2000; Twombly & Holtz, 2008). This article helps to reduce this knowledge void by presenting a general community prevention model (CPM) for reducing the use of HLPs to get high among pre- and early-adolescent children with special application to rural communities.…”