How Many Sexual Partners of an Individual Need to Be Evaluated to Capture HIV/STI Risk Behavior in a Study?

Investigators conducting HIV studies ask participants multiple questions about sexual risk behaviors with their partners to ensure that they can describe the level of HIV risky sexual behavior. The assessment should be as short as possible because of the expense of collecting the data, the burden to the research subject, and ethical concerns. This study used data from the NIMH Collaborative HIV/STD Prevention Trial to answer the question about how many non-spousal/non live-in partners a research participant needs to be asked about to capture sufficient sexual risk behavior (not using a condom with a non-spousal/non live-in partner in the last 3 months). The data provided evidence that 95 % of the sexual risk behavior was captured by asking about two partners while 98 % was captured by three partners. As research funds become increasingly limited, it is important to design as parsimonious and robust a study as possible.

The utility of a composite biological endpoint in HIV/STI prevention trials.

A human immunodeficiency virus (HIV) as a biological endpoint in HIV prevention trials may not be feasible, so investigators have used surrogate biological outcomes. In a multisite trial, the epidemiology of STIs may be different across sites and preclude using one STI as the outcome. This study explored using a composite STI outcome to address that problem. The combined biological endpoint was the incidence of any of six new STIs (chlamydia, gonorrhea, trichomonas (women only), syphilis, herpes simplex virus type 2 infection and HIV) during a 24-month follow up period. We investigated how a composite STI outcome would perform compared to single and dual STI outcomes under various conditions. We simulated outcomes for four populations that represented a wide range of sex and age distributions, and STI prevalences. The simulations demonstrated that a combined biologic outcome was superior to single and dual STI outcomes in assessing intervention effects in 82 % of the cases. A composite biological outcome was effective in detecting intervention effects and might allow more investigations to incorporate multiple biological outcomes in the assessment of behavioral intervention trials for HIV prevention.

Randomized controlled trial of meat compared with multimicronutrient-fortified cereal in infants and toddlers with high stunting rates in diverse settings.

BACKGROUND: Improved complementary feeding is cited as a critical factor for reducing stunting. Consumption of meats has been advocated, but its efficacy in low-resource settings has not been tested. OBJECTIVE: The objective was to test the hypothesis that daily intake of 30 to 45 g meat from 6 to 18 mo of age would result in greater linear growth velocity and improved micronutrient status in comparison with an equicaloric multimicronutrient-fortified cereal. DESIGN: This was a cluster randomized efficacy trial conducted in the Democratic Republic of Congo, Zambia, Guatemala, and Pakistan. Individual daily portions of study foods and education messages to enhance complementary feeding were delivered to participants. Blood tests were obtained at trial completion. RESULTS: A total of 532 (86.1%) and 530 (85.8%) participants from the meat and cereal arms, respectively, completed the study. Linear growth velocity did not differ between treatment groups: 1.00 (95% CI: 0.99, 1.02) and 1.02 (95% CI: 1.00, 1.04) cm/mo for the meat and cereal groups, respectively (P = 0.39). From baseline to 18 mo, stunting [length-for-age z score (LAZ) <-2.0] rates increased from ~33% to nearly 50%. Years of maternal education and maternal height were positively associated with linear growth velocity (P = 0.0006 and 0.003, respectively); LAZ at 6 mo was negatively associated (P < 0.0001). Anemia rates did not differ by group; iron deficiency was significantly lower in the cereal group. CONCLUSION: The high rate of stunting at baseline and the lack of effect of either the meat or multiple micronutrient-fortified cereal intervention to reverse its progression argue for multifaceted interventions beginning in the pre- and early postnatal periods.

The incidence and correlates of symptomatic and asymptomatic Chlamydia trachomatis and Neisseria gonorrhoeae infections in selected populations in five countries.

BACKGROUND: Asymptomatic Chlamydia trachomatis (chlamydia) and Neisseria gonorrhoeae (gonorrhea) infections pose diagnostic and control problems in developing countries. METHODS: Participants in China, India, Peru, Russia, and Zimbabwe were screened for C. trachomatis and N. gonorrhoeae infections and symptoms. RESULTS: A total of 18,014 participants were evaluated at baseline, 15,054 at 12 months, and 14,243 at 24 months. The incidence of chlamydia in men was 2.0 per 100 person years both from baseline to 12 months and from 12 to 24 months, and in women, 4.6 from baseline to 12 months and 3.6 from 12 to 24 months; a range of 31.2% to 100% reported no symptoms across the 5 countries. The incidence of gonorrhea in men was 0.3 per 100 person years both from baseline to 12 months and from 12 to 24 months, and in women, 1.4 from baseline to 12 months and 1.1 from 12 to 24 months; a range of 66.7% to 100% reported no symptoms. Being female, aged 18 to 24 years, and having more than 1 partner were associated with both the infections. In addition, being divorced, separated, or widowed was associated with gonorrhea. Being male, having 6+ years of education, and reporting only 1 partner were associated with having no symptoms among those infected with chlamydia. No variables correlated with asymptomatic gonorrhea among those infected. CONCLUSION: A high prevalence and incidence of asymptomatic sexually transmitted infections was identified among men and women in a wide variety of settings. More effective programs are needed to identify and treat chlamydia and gonorrhea infections, especially among women, young adults, those with multiple partners, those repeatedly infected, and particularly those at risk without symptoms. The risk of transmission from persons with no symptoms requires further study.

High mortality rates for very low birth weight infants in developing countries despite training.

OBJECTIVE: The goal was to determine the effect of training in newborn care and resuscitation on 7-day (early) neonatal mortality rates for very low birth weight (VLBW) infants. The study was designed to test the hypothesis that these training programs would reduce neonatal mortality rates for VLBW infants. METHODS: Local instructors trained birth attendants from 96 rural communities in 6 developing countries in protocol and data collection, the World Health Organization Essential Newborn Care (ENC) course, and a modified version of the American Academy of Pediatrics Neonatal Resuscitation Program (NRP), by using a train-the-trainer model. To test the impact of ENC training, data on infants of 500 to 1499 g were collected by using a before/after, active baseline, controlled study design. A cluster-randomized, controlled trial design was used to test the impact of the NRP. RESULTS: A total of 1096 VLBW (500-1499 g) infants were enrolled, and 98.5% of live-born infants were monitored to 7 days. All-cause, 7-day neonatal mortality, stillbirth, and perinatal mortality rates were not affected by ENC or NRP training. CONCLUSIONS: Neither ENC nor NRP training of birth attendants decreased 7-day neonatal, stillbirth, or perinatal mortality rates for VLBW infants born at home or at first-level facilities. Encouragement of delivery in a facility where a higher level of care is available may be preferable when delivery of a VLBW infant is expected.

Newborn-care training and perinatal mortality in developing countries.

BACKGROUND: Of the 3.7 million neonatal deaths and 3.3 million stillbirths each year, 98% occur in developing countries. An evaluation of community-based interventions designed to reduce the number of these deaths is needed. METHODS: With the use of a train-the-trainer model, local instructors trained birth attendants from rural communities in six countries (Argentina, Democratic Republic of Congo, Guatemala, India, Pakistan, and Zambia) in the World Health Organization Essential Newborn Care course (which focuses on routine neonatal care, resuscitation, thermoregulation, breast-feeding, "kangaroo" [skin-to-skin] care, care of the small baby, and common illnesses) and (except in Argentina) in a modified version of the American Academy of Pediatrics Neonatal Resuscitation Program (which teaches basic resuscitation in depth). The Essential Newborn Care intervention was assessed among 57,643 infants with the use of a before-and-after design. The Neonatal Resuscitation Program intervention was assessed as a cluster-randomized, controlled trial involving 62,366 infants. The primary outcome was neonatal death in the first 7 days after birth. RESULTS: The 7-day follow-up rate was 99.2%. After birth attendants were trained in the Essential Newborn Care course, there was no significant reduction from baseline in the rate of neonatal death from all causes in the 7 days after birth (relative risk with training, 0.99; 95% confidence interval [CI], 0.81 to 1.22) or in the rate of perinatal death; there was a significant reduction in the rate of stillbirth (relative risk with training, 0.69; 95% CI, 0.54 to 0.88; P=0.003). In clusters of births in which attendants had been randomly assigned to receive training in the Neonatal Resuscitation Program, as compared with control clusters, there was no reduction in the rates of neonatal death in the 7 days after birth, stillbirth, or perinatal death. CONCLUSIONS: The rate of neonatal death in the 7 days after birth did not decrease after the introduction of Essential Newborn Care training of community-based birth attendants, although the rate of stillbirths was reduced. Subsequent training in the Neonatal Resuscitation Program did not significantly reduce the mortality rates. (ClinicalTrials.gov number, NCT00136708.)

Prevalence of Sexually Transmitted Diseases and Risk Behaviors from the NIMH Collaborative HIV/STD Prevention Trial.

This cross-sectional study describes the baseline prevalence and correlates of common bacterial and viral sexually transmitted diseases (STDs) and risk behaviors among individuals at high risk for HIV recruited in five low- and middle-income countries. Correlations of risk behaviors and demographic factors with prevalent STDs and the association of STDs with HIV prevalence are examined. Between 2,212 and 5,543 participants were recruited in each of five countries (China, India, Peru, Russia, and Zimbabwe). Standard protocols were used to collect behavioral risk information and biological samples for STD testing. Risk factors for HIV/STD prevalence were evaluated using logistic regression models. STD prevalence was significantly higher for women than men in all countries, and the most prevalent STD was Herpes simplex virus-type 2 (HSV-2). HIV prevalence was generally low (below 5%) except in Zimbabwe (30% among women, 11.7% among men). Prevalence of bacterial STDs was generally low (below 5% for gonorrhea and under 7% for syphilis in all sites), with the exception of syphilis among female sex workers in India. Behavioral and demographic risks for STDs varied widely across the five study sites. Common risks for STDs included female gender, increasing number of recent sex partners, and in some sites, older age, particularly for chronic STDs (i.e., HSV-2 and HIV). Prevalence of HIV was not associated with STDs except in Zimbabwe, which showed a modest correlation between HIV and HSV-2 prevalence (Pearson coefficient = .55). These findings underscore the heterogeneity of global STD and HIV epidemics and suggest that local, focused interventions are needed to achieve significant declines in these infections.

Intracluster correlation adjustments to maintain power in cluster trials for binary outcomes.

Adequately powered sample size calculations for cluster randomized trials primarily depend on the event rate variability, effect size, average cluster size, and intracluster correlation (ICC). Furthermore, an ICC estimate depends on event rate variability among clusters, cluster size, and number of clusters. We evaluated the impact on ICC estimates of event rates, event rate variations, cluster size, and cluster size variations for different numbers of clusters. We also evaluated how the event rate changes at the end of the trial affect ICC estimates. We created one simulation exercise to investigate how different event rates, event rate variations, cluster size, and cluster size variations impact ICC estimates and 95% confidence intervals. A separate simulation exercise in four different trial scenarios examined the impact of an intervention or drug effect in the intervention group on ICC estimates, 95% confidence intervals, and on sample size. The first simulation results suggest that the ICC value depends upon the event rate and event rate variations in addition to the cluster size, cluster size variations, and number of clusters. The second simulation exercise suggested that adjusting the sample size will help to preserve the appropriate power at the end of the trial.

A simulation based technique to estimate intracluster correlation for a binary variable.

Cluster randomized trials have become the design of choice for evaluating the effect of selected interventions on well-known health indicators such as neonatal mortality rate, episiotomy rate, and postpartum hemorrhage rate in a community setting. Determining the sample size of a cluster randomized trial requires a reliable estimate of cluster size and the intracluster correlation (ICC), because sample size can be substantially impacted by these parameters. During the design phase of a trial, the investigators may have estimates of the valid range of the health indicator which is the primary outcome variable. Furthermore, investigators often have an estimate of the average cluster size or range of cluster sizes that exist among the proposed samples they are planning to include in the trial. We present in this article a simulation technique to estimate the ICC value and its distribution for known binary outcome variables and a varying number of clusters and cluster sizes. We applied this technique to estimate ICC values and confidence intervals for a multi-country trial assessing the effect of neonatal resuscitation to decrease seven-day neonatal mortality, where communities within a country were clusters. This simulation technique can be used to estimate the possible ranges of the ICC values and to help to design an appropriately powered trial.

Effect of Hypericum perforatum (St John's wort) in major depressive disorder: a randomized controlled trial.

CONTEXT: Extracts of Hypericum perforatum (St John's wort) are widely used for the treatment of depression of varying severity. Their efficacy in major depressive disorder, however, has not been conclusively demonstrated. OBJECTIVE: To test the efficacy and safety of a well-characterized H perforatum extract (LI-160) in major depressive disorder. DESIGN AND SETTING: Double-blind, randomized, placebo-controlled trial conducted in 12 academic and community psychiatric research clinics in the United States. PARTICIPANTS: Adult outpatients (n = 340) recruited between December 1998 and June 2000 with major depression and a baseline total score on the Hamilton Depression Scale (HAM-D) of at least 20. INTERVENTIONS: Patients were randomly assigned to receive H perforatum, placebo, or sertraline (as an active comparator) for 8 weeks. Based on clinical response, the daily dose of H perforatum could range from 900 to 1500 mg and that of sertraline from 50 to 100 mg. Responders at week 8 could continue blinded treatment for another 18 weeks. MAIN OUTCOME MEASURES: Change in the HAM-D total score from baseline to 8 weeks; rates of full response, determined by the HAM-D and Clinical Global Impressions (CGI) scores. RESULTS: On the 2 primary outcome measures, neither sertraline nor H perforatum was significantly different from placebo. The random regression parameter estimate for mean (SE) change in HAM-D total score from baseline to week 8 (with a greater decline indicating more improvement) was -9.20 (0.67) (95% confidence interval [CI], -10.51 to -7.89) for placebo vs -8.68 (0.68) (95% CI, -10.01 to -7.35) for H perforatum (P =.59) and -10.53 (0.72) (95% CI, -11.94 to -9.12) for sertraline (P =.18). Full response occurred in 31.9% of the placebo-treated patients vs 23.9% of the H perforatum-treated patients (P =.21) and 24.8% of sertraline-treated patients (P =.26). Sertraline was better than placebo on the CGI improvement scale (P =.02), which was a secondary measure in this study. Adverse-effect profiles for H perforatum and sertraline differed relative to placebo. CONCLUSION: This study fails to support the efficacy of H perforatum in moderately severe major depression. The result may be due to low assay sensitivity of the trial, but the complete absence of trends suggestive of efficacy for H perforatum is noteworthy.