J S St Lawrence

Sex work, drug use, HIV infection, and spread of sexually transmitted infections in Moscow, Russian Federation

Rates of HIV-1 infection are growing rapidly, and the epidemic of sexually transmitted infections is continuing at an alarming rate, in the Russian Federation. We did a cross-sectional study of sexually transmitted infections, HIV infection, and drug use in street youth at a juvenile detention facility, adults at homeless detention centres, and women and men at a remand centre in Moscow. 160 (79%) women at the remand centre were sex workers. 91 (51%) homeless women had syphilis. At least one bacterial sexually transmitted infection was present in 97 (58%) female juvenile detainees, 120 (64%) women at the remand centre, and 133 (75%) homeless women. HIV seroprevalence was high in women at the remand centre (n=7 [4%]), adolescent male detainees (5 [3%]), and homeless women (4 [2%]). In view of the interaction between sexually transmitted infections and HIV infection, these findings of high prevalence of sexually transmitted infections show that these disenfranchised populations have the potential to make a disproportionately high contribution to the explosive growth of the HIV epidemic unless interventions targeting these groups are implemented in the Russian Federation.

Factors associated with HIV prevalence in a pre-partum cohort of Zambian women

An ongoing study of mother-to-child human herpes virus-8 (HHV-8) transmission in Zambian women (n = 3160) allowed us to examine the association of medical injections with HIV serostatus while simultaneously accounting for other factors known to be correlated with HIV prevalence. Multi-method data collection included structured interviews, medical record abstraction, clinical examinations, and biological measures. Medically administered intramuscular or intravenous injections in the past five years (but not blood transfusions) were overwhelmingly correlated with HIV prevalence, exceeding the contribution of sexual behaviours in a multivariable logistic regression. Statistically significant associations with HIV also were found for some demographic variables, sexual behaviours, alcohol use, and sexually transmitted diseases (STD). The results confirmed that iatrogenic needle exposure, sexual behaviour, demographic factors, substance use, and STD history are all implicated in Zambian womens HIV+ status. However, the disproportionate association of medical injection history with HIV highlights the need to investigate further and prospectively the role of health-care injection in sub-Saharan Africas HIV epidemic.

Physicians' opinions about partner notification methods: case reporting, patient referral, and provider referral

Background: The United States has relied upon partner notification strategies to help break the chain of infection and re-infection for sexually transmitted diseases (STD). Physicians are a vital link in the system of STD control, but little is known of physician opinions about partner notification strategies. Methods: We collected opinions about partner notification from a national probability sample of physicians in specialties diagnosing STDs. Physicians responded to 17 questions about three relevant forms of STD partner notification: patient based referral, provider based referral, and case reporting. Results: Exploratory factor analyses showed that responses for each form of partner notification could be grouped into four categories: perceived practice norms, infection control, patient relationships, and time/money. Multivariate analyses of the factors showed that physicians endorsed patient based referral most favourably and provider based referral least favourably. Conclusion: Physicians’ opinions about partner notification strategies appear to reflect objective reality in some areas, but not in others. Strategies that improve the fit between physicians’ opinions and effective notification are needed: some are discussed here.

How to contain generalized HIV epidemics? A plea for better evidence to displace speculation.

In the worst generalized HIV epidemics in East and Southern Africa, from one-quarter to three-quarters of women aged 15 years can expect to be living with HIV or to have died with AIDS by age 40 years. This disaster continues in the face of massive HIV prevention programmes based on current inexact knowledge of HIV transmission pathways and risks. To stop this disaster, both the public and public health experts need better information about the specific factors that allow HIV to propagate so extensively in countries with generalized epidemics. This knowledge could be acquired by tracing HIV infections to their source – especially tracing HIV infections in women of all ages, and tracing unexplained HIV infections in children with HIV-negative mothers.

Emotional intimacy predicts condom use: findings in a group at high sexually transmitted disease risk.

Previous studies have reported an inverse relationship between condom use and emotional intimacy. The aim of this study was to determine the relationship between condom use and emotional intimacy. The study was a gonorrhoea case-comparison study with the samples being drawn from public health clinics (cases) and select bars/nightclubs (places) of Houston, TX (n = 215). Data were collected by questionnaires administered on a laptop computer. The majority of respondents were African-American (97.7%), women (69.3%) and had either high school or GED education (72.6%). Condom use with the last sexual partner was analysed along with intimacy with that partner assessed on a 3-point scale. Analysis showed that higher intimacy was related to greater condom use which was significant in men but not in women. In conclusion, these data were opposite to those of previous studies, which showed an inverse relationship between condom use and emotional intimacy. We hypothesize that in a high-risk environment, people exert more effort in protecting those they feel closer to. These data suggest a need to further explore the complex relationship between emotional intimacy and condom use.

Process, efficacy and sample demographics of three approaches to behavioural surveillance for gonorrhoea: case interviews, place surveys, and network studies

We investigated the process and time required to collect 450 interviews in a project to determine the most efficacious behavioural surveillance approaches to detect changes in gonorrhoea prevalence. In total, 150 respondents were recruited in each method. For each of place surveys (bars), gonorrhoea case interviews, and network studies based on seeds from the case and place interviews, we determined the recruitment rate and process. Urine testing for gonorrhoea and chlamydia took place in the place interviews. We present data from Houston, Texas that illustrate the sample characteristics, recruitment rates, and, where appropriate, infection rates. Data indicate that there was high uptake and a rapid recruitment rate from the place surveys, an intermediate rate from the network studies, and that the gonorrhoea case interviews were the most inefficient accrual method for behavioural surveillance. Sample characteristics and biases in each method are described, and conclusions drawn for the relative efficacy of each method for gonorrhoea behavioural surveillance.

Repeating a plea for better research and evidence

cases. Gisselquist et al. call for more evidence on MOTs but they ignore current and reported evidence by UNAIDS and the Global AIDS M&E Team on exactly this evidence. The latter studied MOTs in five generalized high-HIV countries in Africa, with the aim of generating an evidence-based response, allocating resources and developing capacity. The data obtained provide the best evidence yet on this issue. The consensus from these extensive analyses, which are publicly available, is that, on average, over 95% of incident HIV acquisition can be ascribed to sexual intercourse. The MOTs attribute only a minor fraction (less than 1%) of HIV to nosocomial infections (medical injections and transfusion). Furthermore, a model study by Hauri et al. that Gisselquist et al. cite could only attribute 5% of HIV infections to nosocomial factors globally and 2.5% in eastern and southern Africa where HIV is highly prevalent. Contrary to the arguments by Gisselquist et al., there is simply no evidence, whatsoever, for nosocomial infections being responsible, or having any significant role, in the generalized HIV epidemic in sub-Saharan Africa. Secondly, Gisselquist et al. attempt to discredit the large, much acclaimed randomized control trials (RCTs) that have shown MC to be highly effective (approx. 60%) in protecting against heterosexual acquisition of HIV by men, by suggesting that behavioural confounders might be contributing to such findings. In so doing they cite surveys to support their opinions about ‘conflicting evidence’. We know of no empirical or modelled studies negating the principal finding of the African trials. Every systematic review, meta-analysis and the latest Cochrane review of MC studies has confirmed the results of the trials, as well as their methodological validity and power. Gisselquist et al. cite not a single study that questions the results of the trials. In fact, these authors go further by attempting to misconstrue the South African trial. They say that the trial reports an adjusted rate ratio of 1.7 (P 1⁄4 0.09) for ‘HIV incidence associated with a composite “nosocomial risk factor”’. What they fail to report is what the study actually says about this rate ratio, namely that ‘the presence of a nosocomial risk factor was not significantly associated with HIV infection’ and further that when this risk factor was taken into account ‘HIV infection was unchanged’ between the two groups in randomization, with the protective effect of MC being 60% in both. Furthermore, a critical assessment of the observational data from various African countries highlights the need for caution in interpreting the kind of surveys that Gisselquist et al. refer to concerning ‘conflicting evidence’ on the protective role for MC. A prior response by a large consortium of expert researchers to concerns such as those Gisselquist et al. raise is worthy of consultation. While we have no qualms about developing a solid foundation for prevention, including the establishment of rigorous data concerning all possible routes for transmission of HIV, we cannot agree with the dismissal of MC in favour of observational suggestions for MOTs for which no systematic evidence or randomized trials exist and that have already been shown to have relatively little public health significance in the context of the HIV epidemiology. In making their impassioned appeal for evidence, Gisselquist et al. urge a ‘more ambitious goal: to stop generalized epidemics’. While this is not a new goal or proposal, it is not clear as to how these researchers could hope to accomplish this if they attempt to discredit acclaimed studies that have used established scientific norms in providing evidence for major modes of HIV transmission and for what works in prevention in generalized settings. Stopping generalized epidemics cannot be accomplished by targeting the estimated 0.5–2.5% of infections that are nosocomial, as reported in studies of MOTs and by Hauri et al. Rather, population health strategies that contribute to building herd immunity must be the priority. Furthermore, preventing nosocomial infections by interventions in the health service system – already an encouraged practice for a long time – cannot be an HIV-only prevention strategy. Efforts to improve the safety of MC in particular, and the efficiency of the health-care system generally, should assist in ongoing attempts to meet the needs of the population – men, women and children – to reduce the scourge of HIV and improve overall health outcomes. If Gisselquist et al. are unwilling to accept the available evidence, we propose they could do well by conducting well-designed and executed field-based studies on the actual epidemiology of nosocomial infections, as well as cost-effectiveness analyses to help better inform their passionate call.