Childhood sexual abuse and two stages of cigarette smoking in African-American and European-American young women. (September 2016)

• Record Type: Journal Article
• Authors: Sartor, Carolyn E.; Grant, Julia D.; Duncan, Alexis E.; McCutcheon, Vivia V.; Nelson, Elliot C.; Calvert, Wilma J.; Madden, Pamela A.F.; Heath, Andrew C.; Bucholz, Kathleen K.
• Is Part Of: Addictive behaviors. Volume 60(2016)
• Page Start: 131
• Page End: 136
• Publication Date: 2016-09
• Subjects: Smoking -- Sexual abuse -- Women -- African Americans; Substance abuse -- Periodicals; Alcoholism -- Periodicals; Drug addiction -- Periodicals; Nicotine addiction -- Periodicals; Smoking -- Periodicals; Gambling -- Psychological aspects -- Periodicals; Electronic journals; 362.29
• DOI: 10.1016/j.addbeh.2016.03.039
• Languages: English
• ISSNs: 0306-4603

Abstract

Abstract: Objective: The aim of the current study was to determine whether the higher rates of childhood sexual abuse (CSA) but lower rates of cigarette smoking in African-American vs. European-American women can be explained in part by a lower magnitude of association between CSA and smoking in African-American women. Methods: Data were drawn from a same-sex female twin study of substance use (n = 3521; 14.3% African-American). Cox proportional hazards regression analyses using CSA to predict smoking initiation and progression to regular smoking were conducted separately by race/ethnicity. Co-twin status on the smoking outcome was used to adjust for familial influences on smoking (which may overlap with family-level influences on CSA exposure). Results: After adjusting for co-twin status, CSA was associated with smoking initiation in European Americans (hazard ratio (HR) = 1.43, 95% confidence intervals (CI): 1.26–1.62) and with smoking initiation ≤ 16 in African Americans (HR = 1.70, CI: 1.26–2.29). CSA was associated with regular smoking onset ≤ 15 in European Americans (HR = 1.63, CI: 1.21–2.18), with no change in HR after adjusting for co-twin status. In the African-American subsample, the HR for CSA was reduced to non-significance after adjusting for co-twin status (from HR = 3.30, CI: 1.23–8.89 to HR = 1.16, CI: 0.71–1.92 for regular smoking ≤ 15). Conclusions: CSA is associated with moderate elevation in risk for initiating smoking among African-American and European-American women. By contrast, CSA is associated with elevated risk for (adolescent onset) regular smoking only in European-American women. Furthermore, there is significant overlap between risk conferred by CSA and familial influences on regular smoking in African-American but not European-American women. Highlights: Sexual abuse was associated with smoking initiation in both racial/ethnic groups. Sexual abuse predicted adolescent onset regular smoking in European Americans. These associations were independent of familial influences on smoking. Sexual abuse did not independently elevate African Americans' regular smoking risk.