Elements of external validity of tools for health: an intervention for construction laborers

Abstract

Purpose: To examine the external validity of an efficacious tailored smoking cessation and nutrition improvement telephone intervention.

Design: Comparison of characteristics of participants and nonparticipants (representativeness); examination of the extent of intervention implementation.

Setting: Cancer center collaboration with a labor union.

Subjects: Unionized construction laborers.

Intervention: Tailored feedback report, telephone counseling, and supplementary educational materials focused on smoking cessation and improved nutrition.

Measures: Background survey identifying socio-demographic and behavioral characteristics; process evaluation data; and final efficacy survey to determine participant satisfaction.

Analysis: Cross-classification and the chi(2) test of homogeneity were used with categorically measured variables comparing participants and nonparticipants. We compared the means in the two groups for continuously scaled measures using the Student t-test and investigated the multivariable association of the characteristics of participation with a multiple logistic regression. For process data we present frequencies, percentages, and means.

Results: Characteristics associated with participation included self-efficacy to change fruit and vegetable consumption (p = .0009) and self-identification with union's problems (p = 0.05). Eighty-six percent of non-smokers and 61% of smokers completed between 1 and 4 counseling sessions. Over one-half of non-smokers (61%) and smokers (53%) completed 4 or more calls and more smokers (34%) than non-smokers (11%) completed the 5+ sessions.

Conclusions: These results provide a snapshot of characteristics of construction laborers to whom this intervention can be generalized and indicators of feasibility necessary for translating research into practice.