No empirical data support the 54 to 58 mm external diameter of intravaginal rings currently available and in development for contraception and other indications. Understanding how external diameter affects preference, adherence, and acceptability is critical for optimizing future product development.

Our primary objectives were to determine which of 3 nonmedicated rings of differing external diameters was preferred and yielded the highest adherence. Secondary objectives were comparing acceptability, patterns of adherence, and safety of the 3 rings.

In an open-label, 3-way crossover trial, healthy, HIV-uninfected, monogamous, sexually active, nonpregnant, 18-year-old to 40-year-old cisgender women and their male partners in Atlanta, GA and the Bronx, NY were randomly assigned to the sequence of using 3 nonmedicated silicone rings (46 mm, 56 mm, and 66 mm external diameters) continuously for approximately 30 days each (90 days total; November 2021 to December 2022). We tested whether end-of-study preference for any of the 3 rings was greater than 0.33 (binomial proportion, exact test). We used mixed-effect regression models with random intercepts by participants to compare adherence (ring never out for more than 30 minutes in 24 hours) and probability of removals (including reasons for removal), expulsions, and the ring being out of the vagina all day, per ring, per day of use with the 56-mm ring as the reference group and to compare scores on a novel 19-item acceptability scale with items related to ease and experience of use and impact on sex.

Of 24 women, 23 completed the study (median age: 26.7 years [standard deviation: 3.78]). Most were college graduates (92%), White (63%), non-Hispanic (79%), single (79%), and nulliparous (92%). At the end of the study, 59% preferred the smallest (46 mm) ring versus 18% each for the 56-mm and 66-mm rings ( P =.0045). The proportion of participants who were adherent did not differ by ring (46 mm, 78%; 56 mm, 75%; 66 mm, 59%; P =.30); however, odds of expulsion were higher for the 46-mm (odds ratio: 5.72; 95% confidence interval: 1.25–26.1) and 66-mm (odds ratio: 25.9; 95% confidence interval: 6.11–109) rings than for the 56-mm ring. The 66-mm ring also had greater odds of being out (removal or expulsion, any length of time) than the 56-mm ring (odds ratio: 6.50; 95% confidence interval: 3.46–12.2).

Mean acceptability scale scores were identical (4.54/5) for the 46-mm and 56-mm (smallest and medium) rings and significantly higher than for the largest, 66-mm ring (3.94/5; P < .001).

Although the 46-mm ring was preferred by more women and had higher rates of adherence compared to the other 2 rings, the 56-mm ring was rated as equally acceptable with significantly fewer reports of expulsions/removals compared to the other 2 rings. These data confirm that the current 54 to 58 mm diameter of vaginal rings on the market and in development is acceptable and performs well. Future studies should explore the effect of other mechanical attributes, such as compressibility, on preference, adherence, and acceptability, as well as investigating ring characteristics in other populations.