Such multicategorical response questions have been found to have higher agreement with biochemical measures of exposure to nicotine (Campbell, Sanson-Fisher, & Walsh, 2001; Mullen, Carbonari, Tabak, & Glenday, 1991). A total of 4,258 women were screened, of which 725 (17%) were found to be eligible, inhibitor expert including the 360 (49.6%) who were randomly assigned to the three conditions, 120 per group (Figure 1). Eligible women who did not participate included those who declined enrollment, did not show for their appointments, or could not be contacted. Analyses of demographic information collected on screening forms indicated that eligible women who were not enrolled were similar to those who participated in the study with regard to age, race/ethnicity, gestational age at screening, and marital status.

Figure 1. CONSORT diagram indicating randomization and retention. Randomization Eligible women who agreed to participate were scheduled for an appointment at the UCRC. A block randomization method, using blocks of six (two per condition), was used to generate 360 slots, 120 per intervention group (Graziano & Raulin, 1989). Women were not randomized to the study until they presented for the initial assessment. In the case of a missed intake appointment (see Figure 1), the randomized slot was filled by the next woman recruited. Study procedures and measures Women enrolled in the study completed baseline and EOP assessments. At the baseline visit, prior to the conduct of study activities, the women gave informed consent, submitted a saliva sample, and completed the intake questionnaire.

After the assessment, women were escorted to their ultrasound appointment, with the exception of the BP-only group who went directly to the nurse for the BP session. Following the ultrasound, women then attended their MI or BP session. The EOP assessment was conducted in person in the UCRC during the eighth month of pregnancy and consisted of the self-report questionnaire and saliva sample. In addition to demographic and smoking variables, other predictors of smoking cessation were included in the questionnaire. Stage of change was measured using an algorithm developed for pregnancy smoking cessation (Stotts et al., 2002). The algorithm separates individuals into precontemplation, contemplation, preparation, and action based on intentions and behaviors related to quitting smoking during this pregnancy.

Depression was assessed using the Beck Depression Inventory, a widely used self-report measure of depression (Beck, 1967). The variable smoking networks was derived Carfilzomib from one item on the baseline questionnaire: ��How many family members and friends whom you see regularly are smokers?�� Answers were on a Likert scale from 1 = none to 4 = most. Self-reported smoking status was validated by saliva cotinine using a cutoff value of 20 ng/ml. The sample was collected using a cotton dental roll, placed in the mouth for 5�C10 min until saturated.