Therapeutic community-oriented day treatment program for Korean women with alcohol use disorder: a non-randomized pilot feasibility trial

Table 3 Changes in proportion of alcohol abstinence vs. relapse among participants by study condition over 6 months

		Baseline (T0)	3 months (T1)	6 months (T2)	Cochran's Q test (df = 2)^d	Pairwise McNemar tests
		Pre-intervention	Mid-intervention	Post-intervention
		n (%)	n (%)	n (%)
Intervention	Abstinence	13 (68.4%)^a	16 (84.2%)^b	15 (78.9%)^b		T0 and T1, p = 0.453
	Relapse	6 (31.6%)^a	3 (15.8%)^b	4 (21.1%)^b	Q = 1.75, p = 0.417	T1 and T2, p = 1.000
	Total	19 (100%)	19 (100%)	19 (100%)		T0 and T2, p = 0.727
Control	Abstinence	16 (76.2%)^a	5 (23.8%)^a	2 (9.5%)^a		T0 and T1, p = 0.003*
	Relapse	5 (23.8%)^a	16 (76.2%)^a	19 (90.5%)^a	Q = 20.38, p < 0.001	T1 and T2, p = 0.375
	Total	21 (100%)	21 (100%)	21 (100%)		T0 and T2, p < 0.001*

Bold values denote statistically significant results
All participants lost to follow-up were treated as relapses per intention-to-treat principle
Cochran’s Q tested the null hypothesis that the proportions of ‘relapse’ were the same over time
^*McNemar Exact p < 0.017 (Bonferroni correction)
^aBased on participant self-report
^bBased on breath alcohol test

ISSN: 1940-0640