Background: Studies show inconsistent condom use is relatively common among
military personnel. In an effort to increase condom use among soldiers, an intervention
was implemented to examine the effects of condom scent and wrapper graphics on
usage in the Botswana Defence Force (BDF). Current condom use behaviors and
correlates of lower condom use were characterized, and differences in reporting
behaviors from a retrospective survey and a prospective diary were also examined.
xv
Methods: From October 2010–April 2011, 211 sexually-active male BDF soldiers,
aged 18–30, participated in the study. Participants at four military bases received
condoms in a plain blue or camouflage wrapper, either scented or unscented. A
baseline survey collected demographics and HIV risk behaviors. Pre- and postintervention
behaviors were measured with two diaries each. A retrospective survey
was administered immediately after the post-intervention diaries, which queried
participants about behaviors as reported in those diaries. Several participants were also
randomly selected to participate in focus group sessions which included discussions
about condoms, condom use, and reporting of sexual behaviors.
Results: Only 51% of participants reported always using condoms at baseline. Lower
condom use was associated with excessive alcohol use, reported trust for a sexual
partner, and the perception that condoms make sex less enjoyable. Scented condoms
and condoms packaged in a camouflage wrapper were found to increase condom use
rates in the BDF. Most sexual behaviors reported in the survey and diary were similar,
with the exception of behaviors associated with having a regular partner, reports of
exchanging material goods for sex with a casual partner, and reports of some
retrospective condom use categories.
Conclusions: HIV interventions aimed at increasing condom use in the BDF should
address alcohol abuse, beliefs about condoms, and issues of trust. The BDF should
consider providing scented condoms and condoms packaged in a camouflage wrapper,
coupled with messages of correct and consistent condom use, which may help increase
usage. Results suggest that retrospective surveys are useful for measuring recent
xvi
sexual behaviors, although prospective diaries may be more reliable in collecting
routine and sensitive sexual practices in the BDF.