The MIRA trial assessed whether providing diaphragm, lubricant gel, and condoms (intervention) compared with condoms alone (control) could reduce HIV incidence among 5,039 Southern African women. Compared with the control group, the cumulative proportion of last sex acts protected by any method was higher in the intervention group (OR = 1.33; 95% CI 1.18, 1.49); however, only 36.3% of last sex acts were protected by both a male condom and a diaphragm, whereas 36.6% were protected by a diaphragm only. Product substitution (ever deciding to use a diaphragm instead of a condom in the previous 3 months) was reported at every visit by 22.4%, at some visits by 60.7%, and at none of the visits by 16.8% of these women. Women at greater risk for infection through their own or their partner's behavior or who believed the diaphragm protected against HIV were more likely to report product substitution at every visit.