Methods

Study sample

A convenience sample (n = 440) was recruited via advertisements in newspapers and a blog on the website of a condom sales company that directed men to the questionnaire posted on the website of The Kinsey Institute for Research in Sex, Gender, and Reproduction. Inclusion criteria were: male, at least 18 years old, used condoms for PVI in the past 3 months and the ability to read English. The advertisement read as follows: “The Kinsey Institute is needing men to complete a brief, on-line questionnaire on condom use and erection-enhancing substances.” Approval for the study was obtained from the Indiana University Institutional Review Board.

Measures

All assessments focused on the last condom use event for PVI. Men were asked to indicate whether the condom they last used fit “OK” relative to length or whether it was too long or too short. They were also asked whether the width was “OK”, too narrow, or too wide. Men indicating “OK” to both items were classified as having a “good” fit and all others were classified as having “ill-fitting” condoms. Twelve outcomes were assessed (table 1) by items appearing under a stem that read: “Thinking about the last time you used a condom for penile–vaginal sex (PVI), did you…”; response options were “yes” or “no”.

Identification of covariates

In addition to age, three additional covariates were assessed and tested for their relationship with the key variable (ill-fitting condoms): marital status, US resident, and alcohol use during the last time condoms were used for PVI. Marital status was not significantly associated with the key variable (p = 0.17) using a screening criterion of p = 0.10; however, the other three were related to reports of ill-fitting condoms.

Data analysis

Men reporting ill-fitting condoms the last time condoms were used were compared with the remaining men who did not report ill-fit. Bivariate associations were assessed by contingency table analyses employing prevalence ratios, their 95% CI and respective p values. Next, a hierarchical multiple logistic regression model calculated adjusted odds ratios (AOR). The first block used direct entry to control for the covariates. The second block contained only the key variable. The model was used sequentially with the 12 outcomes (table 2).

Results

Characteristics of the sample

Average age was 29.6 years (range 18–67 years, SD 10.1). The majority identified as white (84.3%), with 5.2% identifying as black/African-American, 4.3% identifying as Asian and the remainder identifying as members of other races. Nearly three-quarters (73.9%) lived in the USA; however, a proportion resided in the UK (7.5%) and Canada (3.4%), with the remainder living in 25 different countries. Four men did not answer the questions pertaining to condom length and width, leaving n = 436. Of these, 44.7% reported ill-fitting condoms. The mean age of men reporting ill-fitting condoms was 28.2 years versus 30.7 years for those not reporting ill-fit (t = 2.59, df = 434, p = 0.01). Men reporting US residence (n = 321) were less likely to report ill-fit (42.1%) compared with 115 men reporting residence in other nations (52.2%; p = 0.06). Men reporting alcohol use during last sex (n = 190) were more likely to report ill-fit (50.5%) compared with 246 men not reporting alcohol use (40.2%; p = 0.03).

Table 1 displays bivariate findings comparing the 44.7% of the men reporting ill-fitting condoms with the 55.3% not reporting this problem. As shown, 11 of the 12 outcomes were significantly associated with ill-fitting condom use. With the exception of reports of irritation of the partner's vagina, men reporting ill-fitting condoms were more likely to report breakage, slippage and each of the other nine outcomes. Table 1 also displays the multivariate findings. As shown, the same 11 outcomes remained significant.

Discussion

This is the first study using an event-specific analysis to examine the effect of ill-fitting condoms on several aspects of sexual functioning. Therefore, in addition to observing that men reporting ill-fit were substantially more likely to report breakage and slippage, we found evidence collectively suggesting that ill-fitting condoms may indeed diminish sexual functioning and pleasure during penile–vaginal sex (for both men and women). Moreover, the increased likelihood that men using ill-fitting condoms will remove condoms before sex ends constitutes another form of condom failure. Fortunately, it seems likely that these problems could be rectified through education programmes. Although the findings are limited by the validity of retrospective self-report and the possibility of experiment-wise error (from multiple comparisons), they emphasise the point that men and their female sex partners may benefit from public health efforts designed to promote the improved fit of condoms.