Skip to main content
. 2016 Mar 9;20(12):2812–2826. doi: 10.1007/s10461-016-1344-7

Table 4.

Differences between classes based on sexual behavior (multivariate analysis)

Variable Class 1 Class 2 Class 3 Class 4 Class 5 Wald statistics P value (for difference)
Tested for HIV in the previous 12 months1 (%) 47.4 55.2 56.2 72.7 67.8 9.9 0.042
Ever taken PEP in the past (%) 8.8a 12.8 11.3 28.7a 17.5 9.6 0.047
HIV status of primary partner (%)
 Not in a relationship 62.3 66.8 71.9 63.2 63.5 115.7 <0.0001
 HIV-negative 31.5a,b 15.4a 19.2b 17.3 14.0
 HIV-positive 2.7a,b 1.4c,d 1.1e,f 17.3a,c,e 19.1b,d,f
 Unknown 3.5a 16.4a,b 7.8b 2.2 3.5
Proportion of sexual encounters that happened in saunas (half the time or more) (%) 12.6a 21.6 20.9a 15.0 25.5 22.9 0.0001
Intentional condomless anal sex (%) 10.8a,b,c 46.9a,d,e 0.0b,d,f,g 35.3c,f 17.3e,g 466.1 <0.0001
Reactive HIV test result(%) 1.3 3.4 2.0 3.9 2.1 7.2 0.130
Previously had an STI (%) 31.0a,b,c 50.8a,d 37.9d 64.2b 59.0c 41.5 <0.0001
Attitudes towards HIV testing1 (mean)2 6.67 6.67 6.69 6.72 6.63 11.1 0.026

Letters in superscript (a, b, c, d, e, f, g): proportions and means with the same superscript letter statistically differ at P < 0.005 in post hoc pairwise comparisons (adjusted value for a familywise Type I error rate set at 0.05)

1Post hoc tests did not reveal any significant pairwise mean differences when adjusting the critical value of p to reflect the number of tests we performed

2 Would rather know your HIV status to make the right decisions for your own health and the health of others? Scale varying from 1 “Strongly disagree” to 7 “Strongly agree”