Purpose: Many surveys of adolescent behavior are dependent on self-reported data. We sought to assess the accuracy of adolescent self-report of sexually transmitted diseases (STDs) and pregnancies.
Methods: We randomly selected 149 (118 females, 31 males) adolescents to participate in this accuracy study. Follow-up questionnaires were administered to the 126 patients (99 females, 27 males) who returned after enrollment. Patients were asked about STDs and pregnancies during the follow-up period which ranged from 6 months to 1 year. All patient charts were reviewed to validate post-testing history.
Results: Return visits were made by 126 patients (84%). Fifty-one (40%) denied having had an STD at all during the follow-up period but were found to have had at least one STD. Another 11 (9%) admitted having had an STD but had multiple STDs in excess of what they reported. Only 46% of the patients provided accurate information on the follow-up questionnaire. Of the 99 girls who returned for follow-up, the inaccurate patients (10%) consisted of 9 who reported no visits for a pregnancy but had a pregnancy documented in their charts and 1 who underreported her number of pregnancies. Reliability analysis of the accuracy of STD and pregnancy self-report for our patients showed kappas ranging from 0.185 to 0.413 (slight and fair to moderate, respectively). Pearson correlation coefficients were 0.3107 and 0.4364 for STD and pregnancy, respectively.
Conclusions: Our patients' histories of visits for STDs and pregnancies are often not substantiated by review of their medical records. The reason for the inaccuracies in self-report of sexual behaviors is unclear. Further research in this area should be done. Physicians must confirm patient history concerning sexual practices through appropriate record review and medical evaluation.
PIP: Many studies of adolescent sexual behavior are based on self-reported data. The present study sought to assess the accuracy of adolescents' self-reports of pregnancies and sexually transmitted diseases (STDs). Enrolled were 149 predominantly African-American adolescents (118 females and 31 males) who visited the Children's National Medical Center (Washington, DC) for outpatient care; 126 respondents (99 females and 27 males) completed a follow-up questionnaire 6-12 months later. The medical charts of these 126 adolescents were reviewed to validate post-testing history. Only 46% of adolescents provided accurate information on STDs in the follow-up questionnaire. 51 adolescents denied they had an STD during the follow-up period but, in fact, had at least 1 STD recorded. Another 11 acknowledged having had 1 STD but, in fact, had multiple STDs. In terms of pregnancies, 76% of female participants were accurate responders. 9 females who had a pregnancy recorded in their charts denied a pregnancy in the follow-up questionnaire and 1 underreported the number of pregnancies. Reliability analysis of the accuracy of STD and pregnancy self-reports showed kappas of 0.185 and 0.413, respectively, while the Pearson correlation coefficients were 0.317 and 0.4364, respectively. These findings indicate that adolescents frequently provide inaccurate information regarding their sexual histories. The contribution of factors such as discomfort with the interviewer, a lack of privacy in the interview, and a misunderstanding of information conveyed by health staff about STDs to this phenomenon is unknown, however.