Teenage women's patterns of STD testing and diagnosis may be related to their experience of dating violence.1 One-third of sexually active Massachusetts women participating in the 1999 and 2001 Youth Risk Behavior Surveys had ever been sexually or physically hurt by a date. Some 33% had been tested for an STD, including HIV, and 5% had received a diagnosis. Compared with nonabused women, those reporting only sexual violence were more likely to have been tested for STDs (odds ratio, 1.9), but not to have been infected; those reporting only physical violence had increased odds of having been tested for an STD other than HIV (1.6) and of having had an STD, including HIV (2.2). The odds of testing for STDs other than HIV, of testing for both HIV and other STDs, and of receiving a diagnosis were elevated among women who had experienced both types of violence (2.4-3.0). Commenting on these findings, analysts suggest that women who are sexually abused by a date may have "singular incidents outside of relationships" that lead them to seek testing, whereas those experiencing only physical violence may be "chronically exposed to coercive sex" that they do not perceive as abusive. Women experiencing both types of violence, meanwhile, may have the most contact with medical settings and be the most likely to disclose abuse.

1. Decker MR, Silverman JG and Raj A, Dating violence and sexually transmitted disease/HIV testing and diagnosis among adolescent females, Pediatrics, 2005, 116(2):e272-276, <www.pediatrics.org/cgi/doi,10.1542/peds.2005-0194>, accessed Aug. 11, 2005.


A study using data from an electronic medical database provides "an improved estimate of condom effectiveness" for preventing chlamydia infection.1 Use of the database permitted analysts to examine data on all clients seen at a public STD clinic in Connecticut in 2000-2002; it also permitted them to identify clients who had a partner with chlamydia infection and who thus stood to gain from any benefit of condom use. A total of 1,455 men and women were included in the analysis, of whom 30% reported always using condoms and 11% had an infected partner. Overall, 11% of clients received a chlamydia diagnosis; the proportion was 30% among those with an infected partner, including 13% of consistent condom users and a significantly higher proportion of inconsistent users (34%). In analyses controlling for demographic and behavioral characteristics, the odds of diagnosis were reduced by 90% among clients who always used condoms. The analysts note that the findings may not apply to populations other than STD clinic clients.

1. Niccolai LM et al., Condom effectiveness for prevention of Chlamydia trachomatis infection, Sexually Transmitted Infections, 2005, 81(4):323-325.


In a 2002-2003 survey of San Francisco men who have sex with men, reported levels of recent use of sildenafil (Viagra) were elevated among men who engaged in risky behaviors.1 Overall, 29% of respondents had used the drug within the past six months; the proportion was significantly higher among HIV-positive than among HIV-negative men (39% vs. 27%). Men who engaged in unprotected anal sex with two or more partners, those who did so with an HIV-positive partner and those who went to circuit parties or bathhouses reported higher levels of use than did men who refrained from such behavior. Use also was more common among men who used recreational drugs than among those who did not. The researchers comment that the pattern of associations "strongly suggests that for most participants, there was a nonprescription source for this medication." They urge health care providers to intensify their educational efforts regarding sildenafil use and the pharmaceutical industry to be more circumspect in its promotionof the drug.

1. Paul JP et al., Viagra (sildenafil) use in a population-based sample of U.S. men who have sex with men, Sexually Transmitted Diseases, 2005, 32(9):531-533.


Finnish women who had a levonorgestrel-releasing IUD inserted between 1990 and 1993 had no higher incidence of breast cancer up to 10 years later than did the average 30-54-year-old woman in Finland, according to an analysis comparing data from a large postmarketing study of the device with information from a national cancer registry.1 Data were examined by five-year age-group and by length of time since insertion, and the results uniformly failed to support a causal relationship between use of the IUD and breast cancer. Nevertheless, the analysts note that because of possible bias from nonresponse to the postmarketing survey and because it was impossible to adjust for many characteristics that may influence the risk of breast cancer, "a causal relationship cannot be totally excluded, either."

1. Backman T et al., Use of the levonorgestrel-releasing intrauterine system and breast cancer, Obstetrics & Gynecology, 2005, 106(4):813-817.


HIV-infected teenagers may need help in adhering to regimens of highly active antiretroviral therapy (HAART).1 In 1998, researchers conducting a multisite, longitudinal study of HIV disease progression in adolescents added questions about HAART adherence to the survey instrument. The first time the questions were asked of 231 participants receiving HAART, roughly 60-70% reported adherence. Teenagers with late-stage disease had half as great odds of adherence on the previous nonweekend day as those in the early stage of the disease. The odds of adherence on the previous Saturday and in the previous month were reduced among high school dropouts (odds ratios, 0.6 and 0.5, respectively) and declined as intensity of alcohol use increased (0.7 and 0.8); adherence on the previous Saturday rose with CD4 cell count (1.9). Among 65 teenagers who had initially adhered to their treatment regimen for at least a year, half failed to do so within 12 months; young age and being depressed were associated with an increased likelihood of nonadherence. The researchers write that the findings suggest "an urgent need for better interventions to assist adolescents infected with HIV with their medication regimen."

1. Murphy DA et al., Longitudinal antiretroviral adherence among adolescents infected with human immunodeficiency virus, Archives of Pediatrics and Adolescent Medicine, 2005, 159(8):764-770.


Pregnant women's risk of venous thromboembolism—blood clots forming in deep veins, which may travel to the lungs—appears to begin early in gestation, according to findings from a multisite study of the condition conducted in 2001-2002.1 Of the 5,541 individuals enrolled in the study, 34 were pregnant women and 19 were women who had recently given birth. Nearly half of the pregnant women (15) had developed a deep vein thrombosis during their first trimester. One of the pregnant woman and 14 of the postpartum women had undergone surgery within the three months before diagnosis; eight and one, respectively, had a genetic predisposition to the condition, but for most women in both groups, this was its first occurrence. Pregnant women were hospitalized for an average of 6.4 days to treat acute deep vein thrombosis, and postpartum women were hospitalized for an average of 9.1 days. Noting that prevention and treatment of venous thromboembolism among pregnant women "may reduce maternal morbidity and mortality and improve the outcome of pregnancy," the investigators conclude that "when prophylaxis is indicated, it should be initiated early in pregnancy."

1. James AH, Tapson VF and Goldhaber SZ, Thrombosis during pregnancy and the postpartum period, American Journal of Obstetrics and Gynecology, 2005, 193(1):216-219.


Women who use the pill for as little as six months have a reduced risk of ovarian cancer, but the association applies only to those who discontinued the method because of side effects.1 In a study of ovarian cancer patients at 39 Pennsylvania hospitals in 1994-1998, those who had used the pill once for 1-6 months had 27% lower odds of ovarian cancer than never-users; although the result was similar for women who had given birth, no significant association was found for nulliparous women. The association was entirely due to a reduced risk of disease among women who had quit using the pill within six months because of side effects (odds ratio, 0.6); again, only parous women mirrored the overall result. Women who had used the pill for more than six months had a reduced risk of ovarian cancer regardless of why they stopped using.

1. Greer JB et al., Short-term oral contraceptive use and the risk of epithelial ovarian cancer, American Journal of Epidemiology, 2005, 162(1):66-72.


Three percent of participants in Wave 3 of the National Longitudinal Study of Adolescent Health who provided urine samples for STD testing had trichomoniasis, a frequently asymptomatic infection whose prevalence has not previously been studied in the general population.1 In this nationally representative sample of young adults, whose average age was 22 at the time of the survey, the prevalence of trichomoniasis was significantly higher among women than among men (3% vs. 2%) and was higher in the South than in other regions (3% vs. 1-2%). Hispanic and, especially, black young people had higher rates of infection than whites (2% and 7%, compared with 1%). Among women, the prevalence rate ranged from 1% for whites to 11% for blacks; among men, whites had the lowest rate (1%), and blacks and Native Americans the highest (3-4%). Prevalence also varied by age; overall and for each gender, the oldest respondents had significantly higher rates of infection than the youngest. "Because the potential consequences of this infection are significant," the researchers comment, "greater efforts are needed to reduce the prevalence."

1. Miller WC et al., The prevalence of trichomoniasis in young adults in the United States, Sexually Transmitted Diseases, 2005, 32(10):593-598.


U.S. women who are not proficient in English may miss out on important health screening, according to findings from a multisite study of women making the transition to menopause.1 In yearly study visits between 1997 and 2002, an average of 41-54% of women who spoke only English or who spoke English and another language equally well reported having undergone a Pap test, a clinical breast exam or a mammogram in the previous year; the proportions were 11-12% among those who spoke no English and 3% (for all three tests) among those who spoke another language more fluently than English. In analyses controlling for demographic, socioeconomic and health characteristics, women in the second and third groups were significantly less likely than the most proficient English speakers to have had a Pap test (odds ratios, 0.5-0.6); non-English speakers also had reduced odds of having had a breast exam (0.6). Language proficiency was not associated with women's likelihood of having undergone mammography. The investigators observe that "language barriers are not insurmountable" (as the finding regarding mammography suggests), and stress that "it is time to recognize the role that language barriers play in health disparities and to begin to equip health care providers with the…resources they need to overcome these barriers."

1. Jacobs EA et al., Limited English proficiency and breast and cervical screening in a multiethnic population, American Journal of Public Health, 2005, 95(8):1410-1416.