FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2003350695> ?p ?o ?g. }

• W2003350695 endingPage "82" @default.
• W2003350695 startingPage "79" @default.
• W2003350695 abstract "Weir and Feldblum, and Galavotti et al raise a number of issues regarding our article.1 Our overall intent is not to debate the efficacy of condom use when properly used. Condom efficacy has been demonstrated by Weir,2 di Vincenzi3 and others in prospective studies. Our concern centers on the validity of self-reported condom use in the heterosexual populations that are at highest risk for human immunodeficiency virus incidence. Multiple sexually transmitted disease (STD)/human immunodeficiency virus risk reduction strategies are being evaluated in these high-risk groups. However, almost all of the evaluations are based on self-reported sexual behaviors. We believe that the context of public health clinics may bias the responses—in an STD clinic, there is a socially desirable “correct” (i.e., affirmative) answer to the question, “Do you use condoms?” The potential for reporting bias has been reported in earlier reviews of the subject.4 Almost all evaluations of sexual behavior have used self-report.5 STDs as Outcome Measure We agree with Weir that gonorrhea is the outcome that manifests most rapidly. Weir indicates that condoms may have less protective effects for some of the other STDs. This may be the case for genital herpes and human papilloma virus, in which lesions on extragenital sites can occasionally be seen. However, we have no reason to believe that if condoms were consistently used, they would be any less effective against Trichomonas, Chlamydia, or T. pallidum. Because all subjects were evaluated for disease using the available reference methods, the cases diagnosed at visit two are, in our opinion, incident disease. We reanalyzed our data and found that the proportional contribution of gonorrhea to the diagnoses of STD was constant. At the enrollment visit, 81/199(41%) of those with a diagnosed STD had gonorrhea. At the first follow-up visit, 54/125 (43%) of those subjects with an STD had gonorrhea. Loss to Follow-Up We agree that loss to follow-up could potentially bias the results. In analyzing the data, we considered two different hypotheses for the potential selection bias: Subjects who returned for follow-up may be more compliant, and, therefore, less likely to have an STD and be more likely to give accurate self-reported behavior histories. The bias would be less STD in the follow-up cohort. Alternatively, subjects may be more likely to return if they have a symptomatic STD or are concerned of higher risk exposure, which would bias toward higher STD incidence rates at visit two. In response to specific comments, the TRAC study cohort at enrollment has been described in previous publications.6,7 The specific analyses to evaluate for selection bias were not reported in the paper because of space limitations. We performed gender-stratified analyses that compared subjects with follow-up visits with those lost to follow-up (Table 1). Self-reported condom use patterns were similar. Small differences were found for education and working variables. Women with an STD at visit one were less likely to return, which in our opinion, represents a conservative bias for our outcome of condom use validity.TABLE 1: Gender-Stratified AnalysesTiming and Misclassification of Outcome and Exposure We agree with Weir and Feldblum that subjects infected before the 30-day reporting period, if untreated, could “carry-over” the infection and therefore be misclassified, even if they were fully compliant with safer sex guidelines. From an epidemiologic standpoint, this is unlikely, because the majority of chlamydial and gonococcal infections in men become symptomatic within 30 days of exposure.8 As with most STD clinics, almost all patients are treated presumptively at the initial visit for diagnosable disease such as gonorrhea and trichomoniasis. Approximately 75% of patients with chlamydia are treated presumptively for either non-gonococcal urethritis, mucopurulent cervicitis, or as an STD contact. Specific data from the patients in this study are not available. For gonorrhea, syphilis, and trichomonas, single-dose therapy regimens administered on-site were used, which assured effective treatment. Chlamydia was treated during the study period primarily with doxycycline, 100 mg twice daily for 1 week. We believe that the presence of asymptomatic (as opposed to symptomatic) infection is an irrelevant issue, because all participants received a full complement of reference-grade diagnostic testing at each study visit. Weir and Feldblum charge that the presence of a specific STD diagnosed at visit one and two may represent treatment failure, and therefore should not be included in the analyses. We disagree. First, as indicated, we have reason to believe that the treatments given to the vast majority of patients were effective. Second, there is a large body of data that suggests that gonorrhea and other STD risks, especially for “repeaters” is intimately related to the social network (s) of their sexual partners. In other words, repeated incident STD is representative of high-risk sexual behavior, and the proportion of infected individuals in the partner pool.9,10 When analyzed across all STDs, our findings were consistent. For example, 18/81 (22% subjects with gonorrhea at visit one had gonorrhea diagnosed at visit two, this compared to 36/517 (7%) of the subjects who did not have gonorrhea at enrollment. Similarly, those with chlamydia and trichomonas at visit one were three times likely to have those diseases diagnosed at visit two. For syphilis, the small number of incident cases preclude any analysis. Determination of Condom Use The regression models and predictors that focused on retrospective recall of condom use may be subject to self-report bias. We repeated the logistic regression per Weir's recommendation-eliminating definite STD at visit one as one of the independent variables. The results were essentially unchanged. Both correspondents suggested using coital diaries. In a receptive population, coital diaries may be an alternative, and have been used in a number of settings, such as family planning populations, international settings, and even prostitutes. However, those studies were performed in populations that are fundamentally different from an STD clinic. All studies of coital diaries were performed in populations that were enrolled before an STD diagnosis. In contrast, most STD clinic clients initially present for clinical evaluation related to STD symptoms or contact. Enrolling those individuals who are willing to comply with coital diaries may introduce its own selection bias. Furthermore, STD clients in Baltimore have high rates of substance and alcohol use6 which potentially bias any behavioral data collection scheme. The data that we collected may be as reliable as can be expected in the field setting. Interview Context Galavotti suggested that our results may be due to the interviewers and the interview context. All patients were interviewed by research clinicians who were integrated into the STD clinic operations. These clinicians were extensively trained before participating in the study, and the questionnaire was extensively piloted before use. All results from the study questionnaire were kept confidential and were not included in the medical chart. Nevertheless, we were similarly concerned over the potential for bias in the interview context. In 1992–1993, we repeated the enrollment phase of the study in 305 subjects (TRAC-2), but using dedicated study interviewers. Rates of self-reported sensitive behaviors (drug use, sexual activity, same-sex partners, past history of STDs) were similar in both cohorts, suggesting little, if any interviewer effect. These data are being prepared for publication. We also disagree with Galavotti's assertions on the validity of self-reported data on sensitive behaviors, such as substance abuse and sexual behavior. Outside of highly artificial study situations, these behaviors are difficult to validate. Nevertheless, the literature increasingly indicates that self-administered questionnaire instruments give more valid results. For example, the 1988 National Household Survey of Drug Abuse of 5,018 individuals compared live interviewer telephone to self-administered questionnaire. Reported substance abuse using the self-administered instruments ranged from 33% for marijuana use to 121% higher for cocaine use in the past year.11 A similar study by Aquilino demonstrated the same result.12 Analytic Issues Weir and Feldblum commented on our trichotomization approach to analyze condom use measures. This measure did not display a normal distribution: More than 75% of subjects reported using a condom either “all of the time” or “none of the time” (Table 1). Therefore, we were precluded from using a continuous measure of condom use, and elected to use a more restricted categorization of use, to reduce self-report bias to the extent possible. In summary, we agree that our results may be subject to bias. However, they represent findings collected in a well monitored setting that is representative of the population at highest risk of STD and incident sexually acquired human immunodeficiency virus infection, and where behavioral interventions are needed. However, behavioral interventions are expensive, and the only evaluation tools that are consistently available are based on self-report. As scientists, it is our duty to ensure that these evaluation tools are valid in settings in which they are likely to be used. Condoms have been demonstrated to be effective when used properly and consistently. What the field lacks are valid and reliable means of measuring their use." @default.
• W2003350695 created "2016-06-24" @default.
• W2003350695 creator A5015619058 @default.
• W2003350695 creator A5021858666 @default.
• W2003350695 creator A5030762636 @default.
• W2003350695 creator A5034889103 @default.
• W2003350695 creator A5045328465 @default.
• W2003350695 creator A5059367411 @default.
• W2003350695 date "1996-01-01" @default.
• W2003350695 modified "2023-09-26" @default.
• W2003350695 title "In Response: Condom Use to Prevent Incident STDs" @default.
• W2003350695 cites W1531727731 @default.
• W2003350695 cites W1902006318 @default.
• W2003350695 cites W2010361572 @default.
• W2003350695 cites W2036930640 @default.
• W2003350695 cites W2042060355 @default.
• W2003350695 cites W2048727004 @default.
• W2003350695 cites W2068701798 @default.
• W2003350695 cites W2144237660 @default.
• W2003350695 cites W2302908699 @default.
• W2003350695 cites W2341804666 @default.
• W2003350695 cites W4246870035 @default.
• W2003350695 doi "https://doi.org/10.1097/00007435-199601000-00016" @default.
• W2003350695 hasPublicationYear "1996" @default.
• W2003350695 type Work @default.
• W2003350695 sameAs 2003350695 @default.
• W2003350695 citedByCount "4" @default.
• W2003350695 countsByYear W20033506952012 @default.
• W2003350695 crossrefType "journal-article" @default.
• W2003350695 hasAuthorship W2003350695A5015619058 @default.
• W2003350695 hasAuthorship W2003350695A5021858666 @default.
• W2003350695 hasAuthorship W2003350695A5030762636 @default.
• W2003350695 hasAuthorship W2003350695A5034889103 @default.
• W2003350695 hasAuthorship W2003350695A5045328465 @default.
• W2003350695 hasAuthorship W2003350695A5059367411 @default.
• W2003350695 hasConcept C2776983459 @default.
• W2003350695 hasConcept C2779379456 @default.
• W2003350695 hasConcept C3013748606 @default.
• W2003350695 hasConcept C512399662 @default.
• W2003350695 hasConcept C71924100 @default.
• W2003350695 hasConcept C99454951 @default.
• W2003350695 hasConceptScore W2003350695C2776983459 @default.
• W2003350695 hasConceptScore W2003350695C2779379456 @default.
• W2003350695 hasConceptScore W2003350695C3013748606 @default.
• W2003350695 hasConceptScore W2003350695C512399662 @default.
• W2003350695 hasConceptScore W2003350695C71924100 @default.
• W2003350695 hasConceptScore W2003350695C99454951 @default.
• W2003350695 hasIssue "1" @default.
• W2003350695 hasLocation W20033506951 @default.
• W2003350695 hasOpenAccess W2003350695 @default.
• W2003350695 hasPrimaryLocation W20033506951 @default.
• W2003350695 hasRelatedWork W1976004100 @default.
• W2003350695 hasRelatedWork W2019153873 @default.
• W2003350695 hasRelatedWork W2059356875 @default.
• W2003350695 hasRelatedWork W2088376361 @default.
• W2003350695 hasRelatedWork W2358395877 @default.
• W2003350695 hasRelatedWork W2748952813 @default.
• W2003350695 hasRelatedWork W2764773449 @default.
• W2003350695 hasRelatedWork W2899084033 @default.
• W2003350695 hasRelatedWork W2911663661 @default.
• W2003350695 hasRelatedWork W4255242977 @default.
• W2003350695 hasVolume "23" @default.
• W2003350695 isParatext "false" @default.
• W2003350695 isRetracted "false" @default.
• W2003350695 magId "2003350695" @default.
• W2003350695 workType "article" @default.