Risk Behaviors, Medical Care, and Chlamydial Infection Among Young Men in the United States

American Journal of Public Health|July 2002, Vol 92, No. 71140|Research and Practice|Peer Reviewed|Ku et al.RESEARCH AND PRACTICEObjectives.This study assessed factors related to chlamydial infection among youngmen in the United States.Methods.Data were from interviews of nationally representative samples of 470 menaged 18 to 19 years (teenagers) and 995 men aged 22 to 26 years (young adults) andfrom urine specimens tested by means of polymerase chain reaction.Results.Although a majority of the men reported occasional unprotected intercourse,only a minority perceived themselves to be at risk for contracting a sexually transmit-ted disease (STD). Chlamydial infection was detected in 3.1% of the teenagers and4.5% of the young adults. A minority of those infected had symptoms or had beentested for STDs; very few had been diagnosed with STDs.Conclusions.Chlamydial infection is common but usually asymptomatic and undiag-nosed. Primary and secondary prevention efforts should be increased, particularlyamong young adult men. (Am J Public Health.2002;92:1140–1143)Risk Behaviors, Medical Care, and Chlamydial Infection Among Young Men in the United States|Leighton Ku, PhD, MPH, Michael St. Louis, MD, Carol Farshy, BS, Sevgi Aral, PhD, Charles F. Turner, PhD, Laura D. Lindberg, PhD, and FreyaSonenstein, PhDPrimary prevention of sexually transmitteddiseases (STDs) involves adopting safer sexualbehaviors, whereas secondary prevention in-volves detecting and treating infected personsto reduce transmission to sexual partners. Ourstudy examined Chlamydia trachomatis(re-ferred to here as “chlamydial infection”)among young men for the purpose of makingrecommendations for improving primary andsecondary prevention of the disease. Webased our examination on the National Sur-veys of Adolescent Males (NSAM).METHODSNSAM included two surveys: a 1995 na-tionally representative survey of 1729 menaged 15 to 19 years (which had a 75% re-sponse rate)1and a 1988 nationally represen-tative survey of 1880 men aged 15 to 19years (which had a 74% response rate).2Sev-enty-five percent of the respondents to the1988 survey were reinterviewed in 1995.Data for the teenaged sample were col-lected from 470 men aged 18 to 19 yearswho were part of the 1995 survey; data forthe young adult sample were collected from995 men aged 22 to 26 years who were partof the 1988 survey and who were reinter-viewed in 1995. Both the teenaged and theyoung adult sample weights depend on theoriginal sample probabilities and on poststrati-fication adjustments made to correspond withcensus data. The young adult sample weightsalso include longitudinal adjustments to com-pensate for attrition between 1988 and 1995.Respondents were interviewed about theirbackground, behaviors, and attitudes. Theythen completed self-administered question-naires about sensitive topics, including STDsymptoms. After the interview and question-naire were completed, respondents older than18 years were asked for a urine specimen.They were informed that the specimen wouldbe tested for STDs and not for drugs and thatpositive cases would be reported to health de-partments where legally required.3Respon-dents received $10 to $20 for the interviewand an additional $10 to $20 for the speci-men. Some interviews were conducted bytelephone; in these cases, urine specimenswere not collected.After urine specimens were collected, theywere packed in ice, frozen, and shippedovernight (still packed in ice) to the Centersfor Disease Control and Prevention (CDC) foranalysis. Commercial polymerase chain reac-tion (Amplicor, Roche Diagnostic Systems)was used to test the specimens for chlamydialinfection.4All positive cases were confirmedby ligase chain reaction; there were no discor-dant positives.Laboratory results were not available for allrespondents. Results were unavailable for 382of the 1377 young adult respondents (28%);14% were unavailable for logistical reasons(primarily because interviews were conductedby telephone, but also because of shippingdamage, etc.) and 14% were unavailable be-cause of respondent refusal. Results were un-available for 108 of the 578 teenaged respon-dents (18%), 6% for logistical reasons and12% because of respondent refusal.We conducted extensive analyses to deter-mine whether the missing data caused nonre-sponse biases.5Respondents for whom labo-ratory results were missing were not at higherrisk for chlamydial infection than were thosefor whom results were present. Using multipleimputation methods,6,7we determined thatthe nonresponse bias was negligible. We usedactual laboratory results in our analyses ofthe young adult respondents and teenaged re-spondents. All analyses were weighted andadjusted to take into account the complexsampling design.RESULTSSTD Risk Behaviors and Perceived RiskThree fourths of the teenagers and ninetenths of the young adults surveyed were sex-ually active (Table 1). Three fourths of thesexually active teenagers had engaged in atleast 1 act of unprotected