. Author manuscript; available in PMC: 2011 Jun 1.

Published in final edited form as: J Acquir Immune Defic Syndr. 2010 Jun;54(2):204–214. doi: 10.1097/QAI.0b013e3181d61def

Table 2.

Summary of the Behavioral Outcome Data – Change from Baseline to 24 Months in Proportion of People Reporting Unprotected Sex

Country	No. Venue Pairs	Intervention (I)				Comparison (C)				Difference in Change I-C ³
		No. Participants ¹	People Reporting Unprotected Sex (%) ²			No. Participants ¹	People Reporting Unprotected Sex (%) ²			Difference in Change I-C ³
		No. Participants ¹	Baseline	24 Months	Change	No. Participants ¹	Baseline	24 Months	Change	Point Estimate [95% CI]	p-value
China	20	1130	6.59	4.59	-1.99	1587	5.01	4.32	-0.68	-1.31 [-2.96-0.34]	0.11
India	12	1340	43.46	19.15	-24.32	1474	43.16	16.00	-27.16	2.85 [-0.03-5.85]	0.05
Peru	10	922	54.37	43.95	-10.42	1279	56.79	45.53	-11.26	0.85 [-5.40-6.69]	0.74
Russia⁴	10	690	35.61	28.02	-7.60	730	36.06	32.11	-3.95	-3.65 [-10.40-2.93]	0.23
Zimbabwe	15	2056	21.29	11.86	-9.43	2218	21.30	12.38	-8.92	-0.51 [-5.09-3.48]	0.86

Overall	67	6138	32.27	21.51	-10.75	7288	32.46	22.07	-10.39	-0.36	0.71

Analysis is based on 13426 participants, excluding CPOLs, with non-missing baseline and 24 month information.

Within country and condition, the statistics reported for Baseline, 24 Months and Change are averages over venues with equal weighting for each venue. Across country within condition, the statistics reported are averages over the 5 country averages with equal weighting for each country.

Within country, the point estimate is the average of the difference in venue pair change with equal weighting for each pair. Across country, the point estimate is the average over the 5 country averages with equal weighting for each country. The 95% confidence intervals and the p-values are based on permutation tests.

⁴

Two venue pairs are excluded in Russia due to limited sample sizes at 24 months.