. Author manuscript; available in PMC: 2017 Oct 1.

Published in final edited form as: J Subst Abuse Treat. 2016 Jun 17;69:19–27. doi: 10.1016/j.jsat.2016.06.004

Table 2.

Self-Reported 7-Day Point Prevalence Abstinence (95% Confidence Intervals) at Short and Long Term Follow-ups

	Short term Follow-up			Long Term Follow-up
	Internet Intervention	Specialty Care		Internet Intervention	Specialty Care
	N=205	N=203	p-value	N=205	N=203	p-value
Analysis with missing=smoking^a,^c	16.8% (12.3%, 22.7%)	11.7% (8.0%, 16.8%)	0.14	13.5% (9.4%, 19.0%)	15.6% (11.3%, 21.3%)	0.53
Multiple Imputation Analysis with OR=5^b,^c	17.4% (12.7%, 23.3%)	12.7% (8.7%, 18.0%)	0.19	15.2% (10.7%, 21.3%)	16.9% (12.2%, 22.8%)	0.66

Note:

Analysis treats participants with missing outcome data as smokers.

The MNAR imputation analysis follows the multiple imputation methods proposed in Hedeker et al. (2007). In the imputation model, the odds ratio reflecting the association between missing and smoking was set to 5, indicating a high likelihood that non-responders were smoking.

Model adjusted for sex and presence of psychiatric symptoms, the stratification variables