Participant characteristics associated with withdrawal from a large randomized trial of spermicide effectiveness

Table 3 Association between baseline factors and failure to complete study

	Total	Did not complete study		Relative Risk (95% confidence interval)
	N	n	%
Age
≤25 years	660	317	48.0	1.29 (1.15 – 1.45)
> 25 years	854	318	37.2	1
Relationship
single not living with partner	522	247	47.3	1.21 (1.07 – 1.36)
married or living with partner	992	388	39.1	1
Living children
None	639	288	45.1	1.14 (1.0 1 – 1.28)
Any	875	347	39.7	1
Baseline coital frequency
≥8 acts per month	862	389	45.1	1.21 (1.07 – 1.37)
≤7 acts per month	640	239	37.3	1
Geographic region of US
West	424	215	50.7	1.28 (1.12 – 1.46)
South	460	170	37.0	0.93 (0.80 – 1.09)
Northeast	630	250	39.7	1
Center type
university*	1069	476	44.5	1.25 (1.08 – 1.44)
other	445	159	35.7	1
Recruitment rate at study site
≤4 per month	640	301	47.0	1.23 (1.09 – 1.38)
> 4 per month	874	334	38.2	1
Reimbursement rate
≤$200	462	232	50.2	1.31 (1.16 – 1.47)
>$200	1052	403	38.3	1
Participation in colposcopy study
No	1381	590	42.7	1.26 (0.99, 1.61)^†
Yes	133	45	33.8	1

*University centers were defined as those at which participants were seen in a primary university clinic setting. These included: University of Alabama at Birmingham, Birmingham, AL; University of Tennessee at Memphis, Memphis, TN; The University of Texas Health Science Center at San Antonio, San Antonio, TX; Baylor College of Medicine, Houston, TX; Medical University of South Carolina, Charleston, SC; University of Pittsburgh and the Magee-Womens Research Institute, Pittsburgh, PA; University of Pennsylvania Medical Center, Philadelphia, PA; University of Arizona Health Sciences Center, Tucson, AZ; NYU School of Medicine, New York, NY. Other centers included: Johns Hopkins Medical Services Corporation, Baltimore, MD; Vermont Women's Choice Program of Planned Parenthood, Burlington, VT; Eastern Virginia Medical School, Norfolk, VA; Minneapolis Medical Research Foundation, Minneapolis, MN; Planned Parenthood of Central and Northern Arizona, Phoenix, AZ
^†Although this confidence limit includes 1, the p-value for the association between this factor and early withdrawal was 0.047.

ISSN: 1471-2288