Skip to main content
. Author manuscript; available in PMC: 2019 Jul 6.
Published in final edited form as: AJOB Empir Bioeth. 2019 Apr 19;10(2):79–87. doi: 10.1080/23294515.2019.1593257

Table 3.

Relationship between PREMIS Expectation Variables, Condom Use, and Ring Adherence by Study Visit

PREMIS Expectation Variable Condom Use Adherence to Ringa
No Yes Pb No Yes Pb
Expectation of Personal Benefit
  3-month visit 0.3188 0.8476
   N 499 496 75 543
   Median 70.0 80.0 74.0 70.0
   25th, 75th percentile 50.0, 100.0 50.0, 100.0 50.0, 100.0 50.0, 100.0
  12-month visit 0.3302 0.2049
   N 772 777 138 962
   Median 75.0 80.0 70.0 75.0
   25th, 75th percentile 50.0, 100.0 50.0, 100.0 50.0, 95.0 50.0, 100.0
Expectation of Maximum Aggregate Benefitc
  3-month visit 0.2609 0.9062
   N 418 408 63 439
   Median 70.0 70.0 70.0 70.0
   25th, 75th percentile 50.0, 100.0 50.0, 99.0 50.0, 95.0 50.0, 99.0
  12-month visit 0.3464 0.4730
   N 669 677 120 832
   Median 75.0 75.0 60.0 70.0
   25th, 75th percentile 50.0, 99.0 50.0, 95.0 50.0, 95.0 50.0, 95.0
a

Adherence was defined as “Yes” = residual dapivirine of <23.5 mg and “No” = residual of ≥23.5;

b

P-value from Wilcoxon rank sum test;

c

The higher of the participant’s responses to items 4a and 4b.