Akaike's information criterion in generalized estimating equations

Identifying the intergenerational effects of the 1959-1961 Chinese Great Leap Forward Famine on infant mortality

Using the 1959-1961 Chinese Great Leap Forward Famine as a natural experiment, this study examines the relationship between mothers' prenatal exposure to acute malnutrition and their children's infant mortality risk. According to the results, the effect of mothers' prenatal famine exposure status on children's infant mortality risk depends on the level of famine severity. In regions of low famine severity, mothers' prenatal famine exposure significantly reduces children's infant mortality, whereas in regions of high famine severity, such prenatal exposure increases children's infant mortality although the effect is not statistically significant. Such a curvilinear relationship between mothers' prenatal malnutrition status and their children's infant mortality risk is more complicated than the linear relationship predicted by the original fetal origins hypothesis but is consistent with the more recent developmental origins of health and disease theory.

Monitoring daily affective symptoms and memory function using interactive voice response in outpatients receiving electroconvulsive therapy

OBJECTIVE

Recently, there has been a gradual shift from inpatient-only electroconvulsive therapy (ECT) toward outpatient administration. Potential advantages include convenience and reduced cost. But providers do not have the same opportunity to monitor treatment response and adverse effects as they do with inpatients. This can obviate some of the potential advantages of outpatient ECT, such as tailoring treatment intervals to clinical response. Scheduling is typically algorithmic rather than empirically based. Daily monitoring through an automated telephone, interactive voice response (IVR), is a potential solution to this quandary.

METHODS

To test feasibility of clinical monitoring via IVR, we recruited 26 patients (69% female; mean age, 51 years) receiving outpatient ECT to make daily IVR reports of affective symptoms and subjective memory for 60 days. The IVR also administered a word recognition task daily to test objective memory. Every seventh day, a longer IVR weekly interview included questions about suicidal ideation.

RESULTS

Overall daily call compliance was high (mean, 80%). Most participants (96%) did not consider the calls to be time-consuming. Longitudinal regression analysis using generalized estimating equations revealed that participant objective memory functioning significantly improved during the study (P < 0.05). Of 123 weekly IVR interviews, 41 reports (33%) in 14 patients endorsed suicidal ideation during the previous week.

CONCLUSIONS

Interactive voice response monitoring of outpatient ECT can provide more detailed clinical information than standard outpatient ECT assessment. Interactive voice response data offer providers a comprehensive, longitudinal picture of patient treatment response and adverse effects as a basis for treatment scheduling and ongoing clinical management.

Frequency of metabolic syndrome and 25-hydroxyvitamin D3 levels in patients with non-alcoholic fatty liver disease

AIM

It is known that insulin resistance has an important role in the pathogenesis of non-alcoholic fatty liver disease (NAFLD) and that serum 25-hidroksivitamin D3 [25-(OH)D] levels are found low in the presence of insulin resistance. Metabolic syndrome (MetS) is characterized by insulin resistance. The purpose of the present study was to determine the levels of 25-(OH)D and the frequency of MetS in patients with NAFLD, and to evaluate the association of 25-(OH)D with the histology of NAFLD and metabolic parameters.

METHOD

Sixty-three patients with NAFLD confirmed by liver biopsy (29 females and 34 males, mean age 42.70±9.82 years) and 46 healthy controls (16 females and 30 males, mean age 37.54±8.56 years) were included in the study. International Diabetes Federation criteria were used for MetS diagnosis. Insulin resistance was determined according to the Homeostasis Model of Assessment (HOMA-IR) method. The groups were compared for 25-(OH)D levels and MetS frequencies. Correlation analysis was used to evaluate relationships between 25-(OH)D and metabolic parameters and/or NAFLD histology.

RESULTS

25-(OH)D levels were lower in the NAFLD group compared to the control group (36.06±13.07 ng/mL vs. 51.19±23.45 ng/mL, respectively, P<0.01), while MetS frequency was higher (66.7% vs. 15.2%, P<0.01). In the NAFLD group, 25-(OH)D levels were negatively correlated with non-alcoholic steatohepatitis scores and HOMA-IR (r=-0.317, P=0.011 and r=-0.437, P=0.001, respectively).

CONCLUSION

The present study demonstrated higher frequency of MetS and lower levels of 25-(OH)D in patients with NAFDL, and a negative association of 25-(OH)D levels with non-alcoholic steatohepatitis scores and insulin resistance.

Applications of a Kullback-Leibler Divergence for Comparing Non-nested Models

Wang and Ghosh (2011) proposed a Kullback-Leibler divergence (KLD) which is asymptotically equivalent to the KLD by Goutis and Robert (1998) when the reference model (in comparison with a competing fitted model) is correctly specified and when certain regularity conditions hold true. While properties of the KLD by Wang and Ghosh (2011) have been investigated in the Bayesian framework, this paper further explores the property of this KLD in the frequentist framework using four application examples, each fitted by two competing non-nested models.

Does visual impairment affect mobility over time? The Salisbury Eye Evaluation Study

PURPOSE

To determine if the odds of mobility disability increases at a different rate among visually impaired (VI) as compared with nonvisually impaired (NVI) over an 8-year period.

METHODS

A total of 2520 Salisbury Eye Evaluation Study participants were followed 2, 6, and 8 years after baseline. VI was defined as best-corrected visual acuity worse than 20/40, or visual field of approximately less than 20°. Self-reported difficulty with three tasks was assessed at each visit: walking up 10 steps, walking down 10 steps, and walking 150 feet. Generalized estimating equation models included a 6-year spline, and explored differences in mobility difficulty trajectories by including an interaction between VI status and the spline terms. Odds ratios (OR) and 95% confidence intervals (CI) compared mobility difficulty for each task by VI status.

RESULTS

At baseline, the VI were significantly more likely to report difficulty mobility tasks than the NVI (OR(difficultywalkingup10steps) = 1.37, CI: 1.02-1.80; OR(difficultywalkingdown10steps) = 1.55, CI: 1.16-2.08; OR(difficultywalking150feet) = 1.50, CI: 1.10-2.04). The trajectory of mobility disability did not differ by VI status from baseline to the 6-year visit. However, the difference between the VI and NVI declined at the 8-year visit, which may be due to loss of VI participants at risk of developing mobility difficulty.

CONCLUSIONS

The VI were more likely to report mobility disability than the NVI, but the trajectory of mobility disability was not steeper among the VI as compared to the NVI over the study period.

Randomized community-level HIV prevention intervention trial for men who drink in South African alcohol-serving venues

BACKGROUND

South African alcohol-serving establishments (i.e., shebeens) offer unique opportunities to reduce HIV risks among men who drink.

PURPOSE

To test an individual- and a social structural-level HIV prevention intervention for men who drink in shebeens.

METHODS

Twelve matched pairs of township neighbourhoods were randomized to receive either (i) an HIV prevention intervention (guided by Social Action Theory) to reduce sexual risk and increase risk reduction communication in social networks, or (ii) an attention-matched control intervention that focused on the prevention of relationship violence. At the individual level, the interventions delivered skills building workshops focused on sexual risk reduction. At the social structural level, the intervention aimed to increase conversations about safer sex among men in the shebeens, distributed small media and implemented community educational events. Individual-level outcomes were assessed by following the workshop cohorts for 1 year (N = 984), and community-level outcomes were examined through cross-sectional community surveys conducted for 1 year in the shebeens (N = 9,678).

RESULTS

Men in the HIV prevention workshops demonstrated greater condom use, more HIV prevention-oriented conversations and greater perceptions of safer sex norms than men in the comparison workshops. Changes at the community level demonstrated significant differences in condom use, although the pattern was not consistent over time.

CONCLUSIONS

Multi-level interventions that target men who drink in South African shebeens may help reduce risks for HIV and other sexually transmitted infections.

Augmentative biocontrol in natural marine habitats: persistence, spread and non-target effects of the sea urchin Evechinus chloroticus

Augmentative biocontrol aims to control established pest populations through enhancement of their indigenous enemies. To our knowledge, this approach has not been applied at an operational scale in natural marine habitats, in part because of the perceived risk of adverse non-target effects on native ecosystems. In this paper, we focus on the persistence, spread and non-target effects of the sea urchin Evechinus chloroticus when used as biocontrol agent to eradicate an invasive kelp from Fiordland, New Zealand. Rocky reef macrobenthic assemblages were monitored over 17 months in areas where the indigenous algal canopy was either removed or left intact prior to the translocation of a large number of urchins (>50 ind.·m⁻²). Urchin densities in treated areas significantly declined ∼9 months after transplant, and began spreading to adjacent sites. At the end of the 17-month study, densities had declined to ∼5 ind.·m⁻². Compared to controls, treatment sites showed persistent shifts from kelp forest to urchin barrens, which were accompanied by significant reductions in taxa richness. Although these non-target effects were pronounced, they were considered to be localised and reversible, and arguably outweigh the irreversible and more profound ecological impacts associated with the establishment of an invasive species in a region of high conservation value. Augmentative biocontrol, used in conjunction with traditional control methods, represents a promising tool for the integrated management of marine pests.

One of these things is not like the other: distinctiveness and executive function in preschoolers

There is scant evidence that children younger than 7 years show a memory advantage for distinct information, a memory phenomenon termed the isolation effect (Journal of Experimental Psychology: Learning, Memory, and Cognition, 2001, Vol. 27, pp. 1359-1366). We investigated whether 4-, 5-, and 6-year-olds' developing organizational processing and executive function contributed to the isolation effect, demonstrated when recall was better for a semantically unique target (e.g., sheep, pig, watermelon, duck) rather than a semantically common target (e.g., apple, banana, watermelon, strawberry). To encourage organizational processing, children were asked to categorize each item presented. Children also completed working memory and cognitive flexibility tasks, and only children who scored high in cognitive flexibility demonstrated the isolation effect.

The revolving hospital door: hospital readmissions among patients who are homeless

BACKGROUND

National attention is increasingly focused on hospital readmissions. Little prior research has examined readmissions among patients who are homeless.

OBJECTIVE

The aim of the study was to determine 30-day hospital readmission rates among patients who are homeless and examine factors associated with hospital readmissions in this population.

METHODS

We conducted a retrospective chart review of patients who were homeless and hospitalized at a single urban hospital from May-August 2012. Homelessness was identified by an electronic medical record flag and confirmed by manual chart review. The primary outcome was all-cause hospital readmission to the study hospital within 30 days of hospital discharge. Patient-level and hospitalization-level factors associated with risk for readmission were examined using generalized estimating equations.

RESULTS

There were 113 unique patients who were homeless and admitted to the hospital a total of 266 times during the study period. The mean age was 49 years, 27.4% of patients were women, and 75.2% had Medicaid. Half (50.8%) of all hospitalizations resulted in a 30-day hospital inpatient readmission and 70.3% resulted in either an inpatient readmission, observation status stay, or emergency department visit within 30 days of hospital discharge. Most readmissions occurred early after hospital discharge (53.9% within 1 week, 74.8% within 2 weeks). Discharge to the streets or shelter versus other living situations was associated with increased risk for readmission in multivariable analyses.

CONCLUSIONS

Patients who were homeless had strikingly high 30-day hospital readmission rates. These findings suggest the urgent need for further research and interventions to improve postdischarge care for patients who are homeless.

Impact of postpartum anxiety and depression on child's mental development from two peri-urban communities of Karachi, Pakistan: a quasi-experimental study

BACKGROUND

Postpartum anxiety and depression has detrimental effects on the overall mental development of children. This study aims to assess the impact of postpartum anxiety and depression on children's mental development on all sub-scales in a Pakistani population.

METHODS

A quasi-experimental study was conducted in two peri-urban communities of Karachi, a mega city of Pakistan, to assess the impact of postpartum anxiety and depression on children's growth and mental development. A total of 420 women were enrolled, who had given consent out of 651 pregnant women identified, during February 2004 to December 2005. Data for socio-demographic, home environment and family relationship variables were collected between 36 weeks of pregnancy and within 10 days of childbirth. Mother's levels of anxiety and depression were assessed at 1, 2, 6, 12, 18, 24, and 30 months of childbirth. An indigenous, validated screening instrument- Aga Khan University Anxiety and Depression scale was used and diagnostic confirmation was done through a psychologist's interview, based on DSM IV criteria. Children's growth and development was monitored in the same sequence using an Early Childhood Development tool that consists of five subscales; socio emotional, language, cognitive, gross motor and fine motor development. Physical growth was monitored by measuring height and weight of the child. Data was analyzed using SAS 9.2. Multivariable Generalized Estimating Equations (GEE) logistic regression was conducted to identify association of postpartum anxiety and depression with each early childhood development indicator, adjusting for parental and child factors.

RESULTS

A significant association of postpartum anxiety and depression with delayed development on all five subscales of children's mental development was found in our study. Interestingly, our study found that higher maternal age had adverse effects on child's emotional whereas positive impact on child's cognitive development. Children's stunting had an adverse impact on all five subscales of children's development. Male children were at higher risk for delayed language and gross motor development relative to female children.

CONCLUSIONS

Our study found that postpartum anxiety and depression is associated with adverse outcomes regarding children's mental development on all sub-scales. The impact was accentuated by low family income or child's increasing age.

A population-based analysis of leaving the hospital against medical advice: incidence and associated variables

Background

Prior studies of patients leaving hospital against medical advice (AMA) have been limited by not being population-based or assessing only one type of patient.

Methods

We used administrative data at the Manitoba Centre for Health Policy to evaluate all adult residents of Manitoba, Canada discharged alive from acute care hospitals between April 1, 1990 and February 28, 2009. We identified the rate of leaving AMA, and used multivariable logistic regression to identify socio-demographic and diagnostic variables associated with leaving AMA.

Results

Of 1 916 104 live hospital discharges, 21 417 (1.11%) ended with the patient leaving AMA. The cohort contained 610 187 individuals, of whom 12 588 (2.06%) left AMA once and another 2 986 (0.49%) left AMA more than once. The proportion of AMA discharges did not change over time. Alcohol and drug abuse was the diagnostic group with the highest proportion of AMA discharges, at 11.71%. Having left AMA previously had the strongest association with leaving AMA (odds ratio 170, 95% confidence interval 156–185). Leaving AMA was more common among men, those with lower average household incomes, histories of alcohol or drug abuse or HIV/AIDS. Major surgical procedures were associated with a much lower chance of leaving the hospital AMA.

Conclusions

The rate of leaving hospital AMA did not systematically change over time, but did vary based on patient and illness characteristics. Having left AMA in the past was highly predictive of subsequent AMA events.

Education and age affect skill acquisition and retention in lay rescuers after a European Resuscitation Council CPR/AED course

OBJECTIVES

To examine whether education and age affect skill acquisition and retention in lay rescuers after a European Resuscitation Council (ERC) CPR/AED course.

BACKGROUND

Because of the importance of bystander CPR/AED skills in the setting of cardiac arrest, acquisition and retention of resuscitation skills has gained a great amount of interest.

METHODS

The ERC CPR/AED course format for written and practical evaluation was used. Eighty lay people were trained and evaluated at the end of the course, as well as at one, three, and six months.

RESULTS

Retention of CPR/AED skills improved over time, recording the lowest practical scores at one month after initial training and the lowest written scores at initial training. In practical evaluation scores, when examined longitudinally, age presented a significant adverse effect and higher background education presented a non-significant positive effect. Moreover, regarding written evaluation scores, when examined longitudinally, education presented a significant positive effect while age did not significantly correlate with written scores.

CONCLUSIONS

Education and age affected retention of CPR/AED skills in lay rescuers. Also, our results suggest that the ERC CPR/AED course format may be poorly designed to discriminate between participants with different levels of practical and written resuscitation skills and merit a thorough investigation in future studies.

Beyond buzzing: mosquito watching stimulates malaria bednet use-a household-based cluster-randomized controlled assessor blind educational trial

Malaria remains a severe health problem in Sub-Saharan Africa, with approximately one million deaths and 365 million cases each year. In terms of malaria control, insecticide-treated bednets are an effective tool, and many organizations have distributed free or highly subsidized bednets in malaria endemic areas. Nevertheless, some recipients do not use bednets because of social, environmental or cultural factors. Making vulnerable populations aware of the presence of mosquitoes may improve bednet use among people owning but not using a bednet. We hypothesized that showing freshly collected mosquitoes from the vicinity could improve bednet use in households owning but not using bednets. To test this hypothesis, we applied a household-based cluster-randomized controlled assessor blind educational trial. Indirect observation of mosquitoes, via educational leaflets, produced no change in bednet use, while showing freshly captured mosquitoes led to a 13-fold increase in bednet use. Our results suggest that direct observation of freshly captured mosquitoes can encourage bednet use and may potentially improve effective bednet coverage for malaria control and elimination.

Use of prostate-specific antigen testing as a disease surveillance tool following radical prostatectomy

BACKGROUND

Prostate-specific antigen (PSA) testing is recommended every 6 to 12 months for the first 5 years following radical prostatectomy as a means to detect potential disease recurrence. Despite substantial research on factors affecting treatment decisions, recurrence, and mortality, little is known about whether men receive guideline-concordant surveillance testing or whether receipt varies by year of diagnosis, time since treatment, or other individual characteristics.

METHODS

Surveillance testing following radical prostatectomy among elderly men was examined using Surveillance, Epidemiology, and End Results cancer registry data linked to Medicare claims. Multivariate logistic regression was used to examine the effect of demographic, tumor, and county-level characteristics on the odds of receiving surveillance testing within a given 1-year period following treatment.

RESULTS

Overall, receipt of surveillance testing was high, with 96% of men receiving at least one test the first year after treatment and approximately 80% receiving at least one test in the fifth year after treatment. Odds of not receiving a test declined with time since treatment. Nonmarried men, men with less-advanced disease, and non-Hispanic blacks and Hispanics had higher odds of not receiving a surveillance test. Year of diagnosis did not affect the receipt of surveillance tests.

CONCLUSIONS

Most men receive guideline-concordant surveillance PSA testing after prostatectomy, although evidence of a racial disparity between non-Hispanic whites and some minority groups exists. The decline in surveillance over time suggests the need for well-designed long-term surveillance plans following radical prostatectomy. Cancer 2013;119:3523-3530.. © 2013 American Cancer Society.

The prognostic value of pre-operative and post-operative B-type natriuretic peptides in patients undergoing noncardiac surgery: B-type natriuretic peptide and N-terminal fragment of pro-B-type natriuretic peptide: a systematic review and individual patient data meta-analysis

OBJECTIVES

The objective of this study was to determine whether measuring post-operative B-type natriuretic peptides (NPs) (i.e., B-type natriuretic peptide [BNP] and N-terminal fragment of proBNP [NT-proBNP]) enhances risk stratification in adult patients undergoing noncardiac surgery, in whom a pre-operative NP has been measured.

BACKGROUND

Pre-operative NP concentrations are powerful independent predictors of perioperative cardiovascular complications, but recent studies have reported that elevated post-operative NP concentrations are independently associated with these complications. It is not clear whether there is value in measuring post-operative NP when a pre-operative measurement has been done.

METHODS

We conducted a systematic review and individual patient data meta-analysis to determine whether the addition of post-operative NP levels enhanced the prediction of the composite of death and nonfatal myocardial infarction at 30 and ≥180 days after surgery.

RESULTS

Eighteen eligible studies provided individual patient data (n = 2,179). Adding post-operative NP to a risk prediction model containing pre-operative NP improved model fit and risk classification at both 30 days (corrected quasi-likelihood under the independence model criterion: 1,280 to 1,204; net reclassification index: 20%; p < 0.001) and ≥180 days (corrected quasi-likelihood under the independence model criterion: 1,320 to 1,300; net reclassification index: 11%; p = 0.003). Elevated post-operative NP was the strongest independent predictor of the primary outcome at 30 days (odds ratio: 3.7; 95% confidence interval: 2.2 to 6.2; p < 0.001) and ≥180 days (odds ratio: 2.2; 95% confidence interval: 1.9 to 2.7; p < 0.001) after surgery.

CONCLUSIONS

Additional post-operative NP measurement enhanced risk stratification for the composite outcomes of death or nonfatal myocardial infarction at 30 days and ≥180 days after noncardiac surgery compared with a pre-operative NP measurement alone.

Predicting acute and persistent neuropathy associated with oxaliplatin

OBJECTIVES

We sought to predict oxaliplatin-associated peripheral neuropathy during modified FOLFOX6 (mFOLFOX6) therapy.

METHODS

Equal numbers of male and female patients with previously untreated, primary or recurrent colorectal cancer were followed through a first course of mFOLFOX6 with 85 mg/m² oxaliplatin every 2 weeks. Accounting for correlation among a subject's cycle, logistic regression estimated per cycle risk of acute (under 14 d) and persistent (14 d or more) neuropathy. Proportional hazards regression predicted time to persistent neuropathy.

RESULTS

Among mFOLFOX6 recipients (n = 50, age 58.9 ± 10.1 y), 36% received concomitant bevacizumab. Of the total number of cycles, 94.2% (422/448) were evaluable. Most (84%) subjects reported neuropathy at least once; 74% reported acute and 48% reported persistent symptoms. On multivariate analysis, risk factors shared by acute and persistent neuropathy were body surface area >2.0, acute neuropathy in a past cycle, and lower body weight. In addition, risk of acute neuropathy decreased with age (adjusted for renal function and winter season), whereas risk of persistent neuropathy increased with cumulative dose of oxaliplatin and persistent neuropathy in a past cycle. Concomitant bevacizumab was not a risk factor when administered in stage IV disease but was associated with persistent neuropathy when administered experimentally in stage III. Females had no increased risk of either form of neuropathy. After 3 cycles, weight, body surface area, and prior acute neuropathy predicted time to persistent neuropathy.

CONCLUSIONS

Routinely available clinical factors predict acute and persistent neuropathy associated with oxaliplatin. When validated, the proposed prognostic score for persistent neuropathy can help clinicians counsel patients about chemotherapy.

Spatial release from masking improves sound pattern discrimination along a biologically relevant pulse-rate continuum in gray treefrogs

Many frogs form large choruses during their mating season in which males produce loud advertisement calls to attract females and repel rival males. High background noise levels in these social aggregations can impair vocal perception. In humans, spatial release from masking contributes to our ability to understand speech in noisy social groups. Here, we tested the hypothesis that spatial separation between target signals and 'chorus-shaped noise' improves the ability of female gray treefrogs (Hyla chrysoscelis) to perform a behavioral discrimination task based on perceiving differences in the pulsatile structure of advertisement calls. We used two-stimulus choice tests to measure phonotaxis (approach toward sound) in response to calls differing in pulse rate along a biologically relevant continuum between conspecific (50 pulses s(-1)) and heterospecific (20 pulses s(-1)) calls. Signals were presented in quiet, in colocated noise, and in spatially separated noise. In quiet conditions, females exhibited robust preferences for calls with relatively faster pulse rates more typical of conspecific calls. Behavioral discrimination between calls differing in pulse rate was impaired in the presence of colocated noise but similar between quiet and spatially separated noise conditions. Our results indicate that spatial release from energetic masking facilitates a biologically important temporal discrimination task in frogs. We discuss these results in light of previous work on spatial release from masking in frogs and other animals.

Dredging displaces bottlenose dolphins from an urbanised foraging patch

The exponential growth of the human population and its increasing industrial development often involve large scale modifications of the environment. In the marine context, coastal urbanisation and harbour expansion to accommodate the rising levels of shipping and offshore energy exploitation require dredging to modify the shoreline and sea floor. While the consequences of dredging on invertebrates and fish are relatively well documented, no study has robustly tested the effects on large marine vertebrates. We monitored the attendance of common bottlenose dolphins (Tursiops truncatus) to a recently established urbanised foraging patch, Aberdeen harbour (Scotland), and modelled the effect of dredging operations on site usage. We found that higher intensities of dredging caused the dolphins to spend less time in the harbour, despite high baseline levels of disturbance and the importance of the area as a foraging patch.

Denitrification and potential nitrous oxide and carbon dioxide production in brownfield wetland soils

Brownfields, previously developed sites that are derelict, vacant, or underused, are ubiquitous in urban areas. Wetlands on brownfields often retain rain and stormwater longer than the surrounding landscape because they are low-lying; this increases the possibility for these areas to process waterborne contaminants from the urban environment. In the northeastern United States, atmospheric deposition of nitrate (NO) is high. Denitrification, a microbial process common in wetlands, is a means of removing excess NO. Nitrogen gas is the desired end product of denitrification, but incomplete denitrification results in the production of NO, a greenhouse gas. The goal of this study was to investigate the potential of brownfield wetlands to serve as sinks for inorganic nitrogen and sources of greenhouse gases. We examined limitations to denitrification and NO production in brownfield wetland soils in New Jersey. Soil C:N ratios were high (18-40) and intact core denitrification (-0.78 to 11.6 μg NO-N kg dry soil d) and N mineralization (0.11-2.97 mg N kg dry soil d) were low for all sites. However, soil NO increased during dry periods. Nitrate additions to soil slurries increased denitrification rates, whereas labile C additions did not, indicating that soil denitrifiers were nitrogen limited. Incubations indicated that the end product of denitrification was primarily NO and not N. These results indicate that brownfield wetlands can develop significant denitrification capacity, potentially causing NO limitation. They might be significant sinks for atmospheric NO but may also become a significant source of NO if NO deposition were to increase.

Model selection of generalized estimating equations with multiply imputed longitudinal data

Longitudinal data often encounter missingness with monotone and/or intermittent missing patterns. Multiple imputation (MI) has been popularly employed for analysis of missing longitudinal data. In particular, the MI-GEE method has been proposed for inference of generalized estimating equations (GEE) when missing data are imputed via MI. However, little is known about how to perform model selection with multiply imputed longitudinal data. In this work, we extend the existing GEE model selection criteria, including the "quasi-likelihood under the independence model criterion" (QIC) and the "missing longitudinal information criterion" (MLIC), to accommodate multiple imputed datasets for selection of the MI-GEE mean model. According to real data analyses from a schizophrenia study and an AIDS study, as well as simulations under nonmonotone missingness with moderate proportion of missing observations, we conclude that: (i) more than a few imputed datasets are required for stable and reliable model selection in MI-GEE analysis; (ii) the MI-based GEE model selection methods with a suitable number of imputations generally perform well, while the naive application of existing model selection methods by simply ignoring missing observations may lead to very poor performance; (iii) the model selection criteria based on improper (frequentist) multiple imputation generally performs better than their analogies based on proper (Bayesian) multiple imputation.

Stimulant treatment for attention-deficit hyperactivity disorder and risk of developing substance use disorder

BACKGROUND

Attention-deficit hyperactivity disorder (ADHD) is linked to increased risk for substance use disorders and nicotine dependence.

AIMS

To examine the effects of stimulant treatment on subsequent risk for substance use disorder and nicotine dependence in a prospective longitudinal ADHD case-control study.

METHOD

At baseline we assessed ADHD, conduct disorder and oppositional defiant disorder. Substance use disorders, nicotine dependence and stimulant treatment were assessed retrospectively after a mean follow-up of 4.4 years, at a mean age of 16.4 years.

RESULTS

Stimulant treatment of ADHD was linked to a reduced risk for substance use disorders compared with no stimulant treatment, even after controlling for conduct disorder and oppositional defiant disorder (hazard ratio (HR) = 1.91, 95% CI 1.10-3.36), but not to nicotine dependence (HR = 1.12, 95% CI 0.45-2.96). Within the stimulant-treated group, a protective effect of age at first stimulant use on substance use disorder development was found, which diminished with age, and seemed to reverse around the age of 18.

CONCLUSIONS

Stimulant treatment appears to lower the risk of developing substance use disorders and does not have an impact on the development of nicotine dependence in adolescents with ADHD.

Forecasting daily emergency department visits using calendar variables and ambient temperature readings

OBJECTIVES

This study aimed to develop different models to forecast the daily number of patients seeking emergency department (ED) care in a general hospital according to calendar variables and ambient temperature readings and to compare the models in terms of forecasting accuracy.

METHODS

The authors developed and tested six different models of ED patient visits using total daily counts of patient visits to an ED in Sao Paulo, Brazil, from January 1, 2008, to December 31, 2010. The first 33 months of the data set were used to develop the ED patient visits forecasting models (the training set), leaving the last 3 months to measure each model's forecasting accuracy by the mean absolute percentage error (MAPE). Forecasting models were developed using three different time-series analysis methods: generalized linear models (GLM), generalized estimating equations (GEE), and seasonal autoregressive integrated moving average (SARIMA). For each method, models were explored with and without the effect of mean daily temperature as a predictive variable.

RESULTS

The daily mean number of ED visits was 389, ranging from 166 to 613. Data showed a weekly seasonal distribution, with highest patient volumes on Mondays and lowest patient volumes on weekends. There was little variation in daily visits by month. GLM and GEE models showed better forecasting accuracy than SARIMA models. For instance, the MAPEs from GLM models and GEE models at the first month of forecasting (October 2012) were 11.5 and 10.8% (models with and without control for the temperature effect, respectively), while the MAPEs from SARIMA models were 12.8 and 11.7%. For all models, controlling for the effect of temperature resulted in worse or similar forecasting ability than models with calendar variables alone, and forecasting accuracy was better for the short-term horizon (7 days in advance) than for the longer term (30 days in advance).

CONCLUSIONS

This study indicates that time-series models can be developed to provide forecasts of daily ED patient visits, and forecasting ability was dependent on the type of model employed and the length of the time horizon being predicted. In this setting, GLM and GEE models showed better accuracy than SARIMA models. Including information about ambient temperature in the models did not improve forecasting accuracy. Forecasting models based on calendar variables alone did in general detect patterns of daily variability in ED volume and thus could be used for developing an automated system for better planning of personnel resources.

Diarrhoea in the ICU: respective contribution of feeding and antibiotics

Introduction

Diarrhoea is frequently reported in the ICU. Little is known about diarrhoea incidence and the role of the different risk factors alone or in combination. This prospective observational study aims at determining diarrhoea incidence and risk factors in the first 2 weeks of ICU stay, focusing on the respective contribution of feeding, antibiotics, and antifungal drugs.

Methods

Out of 422 patients consecutively admitted into a mixed medical–surgical ICU during a 2-month period, 278 patients were included according to the following criteria: ICU stay >24 hours, no admission diagnosis of gastrointestinal bleeding, and absence of enterostomy or colostomy. Diarrhoea was defined as at least three liquid stools per day. Diarrhoea episodes occurring during the first day in the ICU, related to the use of laxative drugs or Clostridium difficile infection, were not analysed. Multivariate and stratified analyses were performed to determine diarrhoea risk factors, and the impact of the combination of enteral nutrition (EN) with antibiotics or antifungal drugs.

Results

A total of 1,595 patient-days were analysed. Diarrhoea was observed in 38 patients (14%) and on 83 patient-days (incidence rate: 5.2 per 100 patient-days). The median day of diarrhoea onset was the sixth day, and 89% of patients had ≤4 diarrhoea days. The incidence of C. difficile infection was 0.7%. Diarrhoea risk factors were EN covering >60% of energy target (relative risk = 1.75 (1.02 to 3.01)), antibiotics (relative risk = 3.64 (1.26 to 10.51)) and antifungal drugs (relative risk = 2.79 (1.16 to 6.70)). EN delivery per se was not a diarrhoea risk factor. In patients receiving >60% of energy target by EN, diarrhoea risk was increased by the presence of antibiotics (relative risk = 4.8 (2.1 to 13.7)) or antifungal drugs (relative risk = 5.0 (2.8 to 8.7)).

Conclusion

Diarrhoea incidence during the first 2 weeks in a mixed population of patients in a tertiary ICU is 14%. Diarrhoea risk factors are EN covering >60% of energy target, use of antibiotics, and use of antifungal drugs. The combination of EN covering >60% of energy target with antibiotics or antifungal drugs increases the incidence of diarrhoea.

A daily process examination of the bidirectional relationship between craving and alcohol consumption as measured via interactive voice response

BACKGROUND

Craving is a central component to alcohol use disorders, although there are contradictory findings in the literature regarding the importance of craving in alcohol use. The study goal was to examine the bidirectional relationship between craving and alcohol consumption in heavy drinkers.

METHODS

Participants received brief alcohol interventions from their primary care physicians and then were asked to make daily reports of craving and alcohol consumption to an interactive voice response (IVR) telephone system for 180 days. The study sample included 246 participants (166 men) with mean age of 46. Ninety-seven percent were Caucasian and 66% met criteria for alcohol dependence. Analysis used generalized estimating equations (GEE) to evaluate whether craving intensity predicted next day alcohol consumption and whether alcohol consumption predicted next day craving intensity. Significant interactions with gender led to stratified analyses.

RESULTS

GEE analyses revealed a significant bidirectional relationship between craving and drinking, where craving intensity predicted next day total drinks consumed (p = 0.001), and total drinks predicted next day craving intensity (p = 0.02). Exploratory analysis found that gender significantly moderated the craving-drinking relationship (p = 0.002) with men increasing next day alcohol use more (b = 0.19) than women (b = 0.08).

CONCLUSIONS

Findings suggest a bidirectional relationship between craving and drinking may contribute to the development or maintenance of heavy drinking, particularly for men. Based on our findings, we recommend that during brief interventions, physicians address both drinking and craving and provide advice for coping with craving.

Safety in numbers: are major cities the safest places in the United States?

STUDY OBJECTIVES

Many US cities have experienced population reductions, often blamed on crime and interpersonal injury. Yet the overall injury risk in urban areas compared with suburban and rural areas has not been fully described. We begin to investigate this evidence gap by looking specifically at injury-related mortality risk, determining the risk of all injury death across the rural-urban continuum.

METHODS

A cross-sectional time-series analysis of US injury deaths from 1999 to 2006 in counties classified according to the rural-urban continuum was conducted. Negative binomial generalized estimating equations and tests for trend were completed. Total injury deaths were the primary comparator, whereas differences by mechanism and age were also explored.

RESULTS

A total of 1,295,919 injury deaths in 3,141 US counties were analyzed. Injury mortality increased with increasing rurality. Urban counties demonstrated the lowest death rates, significantly less than rural counties (mean difference=24.0 per 100,000; 95% confidence interval 16.4 to 31.6 per 100,000). After adjustment, the risk of injury death was 1.22 times higher in the most rural counties compared with the most urban (95% confidence interval 1.07 to 1.39).

CONCLUSION

Using total injury death rate as an overall safety metric, US urban counties were safer than their rural counterparts, and injury death risk increased steadily as counties became more rural. Greater emphasis on elevated injury-related mortality risk outside of large cities, attention to locality-specific injury prevention priorities, and an increased focus on matching emergency care needs to emergency care resources are in order.

Predictive model for surgical site infection risk after surgery for high-energy lower-extremity fractures: development of the risk of infection in orthopedic trauma surgery score

BACKGROUND

Current infection risk scores are not designed to predict the likelihood of surgical site infection after orthopedic fracture surgery. We hypothesized that the National Nosocomial Infections Surveillance (NNIS) System and the Study on the Efficacy of Nosocomial Infection Control (SENIC) scores are not predictive of infection after orthopedic fracture surgery and that risk factors for infection can be identified and a new score created (Emerg Infect Dis. 2003;9:196-203).

METHODS

We conducted a secondary analysis of data from a trial involving internal fixation of 235 tibial plateau, pilon, and calcaneus fractures treated between 2007 and 2010 at a Level I trauma center. The predictive value of the NNIS System and SENIC scores was evaluated based on areas under the receiver operating characteristic (ROC) curve. Bivariate and multiple logistic regression analyses were used to build an improved prediction model, creating the Risk of Infection in Orthopedic Trauma Surgery (RIOTS) score. The predictive value of the RIOTS score was evaluated via the ROC curve.

RESULTS

NNIS System and SENIC scores were not predictive of surgical site infection after orthopedic fracture surgery. In our final regression model, the relative odds of infection among patients with AO [Arbeitsgemeinschaft für Osteosynthesefragen] type C3 or Sanders type 4 fractures compared with fractures of lower classification was 5.40. American Society of Anesthesiologists class 3 or higher and body mass index less than 30 were also predictive of infection, with odds ratios of 2.87 and 3.49, respectively. The area under the ROC curve for the RIOTS score was 0.75, significantly higher than the areas for the NNIS System and SENIC scores.

CONCLUSION

The NNIS System and SENIC scores were not useful in predicting the risk of infection after fixation of fractures. We propose a new score that incorporates fracture classification, American Society of Anesthesiologists classification, and body mass index as predictors of infection.

LEVEL OF EVIDENCE

Prognostic study, level II.

Rationale and design of the Sodium Lowering In Dialysate (SoLID) trial: a randomised controlled trial of low versus standard dialysate sodium concentration during hemodialysis for regression of left ventricular mass

BACKGROUND

The current literature recognises that left ventricular hypertrophy makes a key contribution to the high rate of premature cardiovascular mortality in dialysis patients. Determining how we might intervene to ameliorate left ventricular hypertrophy in dialysis populations has become a research priority. Reducing sodium exposure through lower dialysate sodium may be a promising intervention in this regard. However there is clinical equipoise around this intervention because the benefit has not yet been demonstrated in a robust prospective clinical trial, and several observational studies have suggested sodium lowering interventions may be deleterious in some dialysis patients.

METHODS/DESIGN

The Sodium Lowering in Dialysate (SoLID) study is funded by the Health Research Council of New Zealand. It is a multi-centre, prospective, randomised, single-blind (outcomes assessor), controlled parallel assignment 3-year clinical trial. The SoLID study is designed to study what impact low dialysate sodium has upon cardiovascular risk in dialysis patients. The study intends to enrol 118 home hemodialysis patients from 6 sites in New Zealand over 24 months and follow up each participant over 12 months. Key exclusion criteria are: patients who dialyse more frequently than 3.5 times per week, pre-dialysis serum sodium of <135 mM, and maintenance hemodiafiltration. In addition, some medical conditions, treatments or participation in other dialysis trials, which contraindicate the SoLID study intervention or confound its effects, will be exclusion criteria. The intervention and control groups will be dialysed using dialysate sodium 135 mM and 140 mM respectively, for 12 months. The primary outcome measure is left ventricular mass index, as measured by cardiac magnetic resonance imaging, after 12 months of intervention. Eleven or more secondary outcomes will be studied in an attempt to better understand the physiologic and clinical mechanisms by which lower dialysate sodium alters the primary end point.

DISCUSSION

The SoLID study is designed to clarify the effect of low dialysate sodium upon the cardiovascular outcomes of dialysis patients. The study results will provide much needed information about the efficacy of a cost effective, economically sustainable solution to a condition which is curtailing the lives of so many dialysis patients.

TRIAL REGISTRATION

Australian and New Zealand Clinical Trials Registry number: ACTRN12611000975998.

Multilevel analysis of bacterial counts from chronic periodontitis after root planing/scaling, surgery, and systemic and local antibiotics: 2-year results

AIM

To follow changes (over 2 years) in subgingival bacterial counts of five microbial complexes including health-related Actinomyces spp. in deeper pockets (≥5 mm) after periodontal treatments.

METHODS

EIGHT DIFFERENT TREATMENTS WERE STUDIED: (1) scaling+root planing (SRP); (2) periodontal surgery (SURG)+systemic amoxicillin (AMOX)+systemic metronidazole (MET); (3) SURG+locally delivered tetracycline (TET); (4) SURG; (5) AMOX+MET+TET; (6) AMOX+MET; (7) TET; and (8) SURG+AMOX+MET+TET. Antibiotics were given immediately following SRP. Subgingival plaque was collected mesiobuccally from each tooth, except third molars, from 176 subjects, completing the study, at baseline, 3, 6, 12, 18, and 24 months post-treatment and analysed for 40 different bacteria using checkerboard hybridization. A negative binomial (NB) generalized estimating equation (NB GEE) model was used to analyze count data and a logistic GEE was used for proportions.

RESULTS

We observed short-term beneficial changes in the composition of the red complex of up to 3 months by treating subjects with AMOX+MET+TET. Similar short-term improvements with the same treatment were observed for Tannerella forsythia and Treponema denticola of the red complex. SURG had also short-term beneficial effect on Porphyromonas gingivalis. No periodontal treatments applied to severely affected sites promoted the growth of Actinomyces. Smoking elevated counts of both the red and orange complex while bleeding on probing (BOP) and gingival redness were also predictors of more red complex counts. Comparatively similar findings were obtained by analyzing counts and by analyzing proportions.

CONCLUSIONS

Although short-term reductions in the counts of the red complex were observed in sites that were treated with AMOX+MET+TET, long-term significant effects were not observed with any of the eight treatments. Poor oral hygiene in patients with severe chronic periodontitis diminished the beneficial effects of treatment.

Caffeine intake and risk of basal cell and squamous cell carcinomas of the skin in an 11-year prospective study

PURPOSE

Caffeine may repair skin damage induced by excessive exposure to ultraviolet light. The purpose of this study was to investigate the association between caffeine intake and incidence of basal cell (BCC) and squamous cell carcinoma (SCC). We also assessed the associations between coffee consumption and incidence of these skin cancers.

METHODS

Caffeine intake and consumption of coffee were estimated from food frequency questionnaires assessed in 1992, 1994, and 1996 among 1,325 randomly selected adult residents of a subtropical Australian community. All histologically confirmed tumours of BCC and SCC occurring between 1997 and 2007 were recorded. Associations with BCC and SCC were assessed using Poisson and negative binomial regression models and were adjusted for confounders including skin type and indicators of past sun exposure.

RESULTS

There was no association between total caffeine intake and incidence of BCC or SCC. Participants with prior skin cancers, however, had a 25% lower risk of BCC if they were in the highest tertile of total caffeine intake (equivalent to daily consumption of four cups of regular coffee) compared with the lowest tertile (multivariable RR 0.75; 95% CI 0.57-0.97, P trend = 0.025). There was no dose-response relationship with SCC. Consumption of neither caffeinated nor decaffeinated coffee was associated with BCC or SCC.

CONCLUSIONS

Among people with prior skin cancers, a relatively high caffeine intake may help prevent subsequent BCC development. However, caffeine intake appears not to influence the risk of SCC.

Parasite-induced and parasite development-dependent alteration of the swimming behavior of fish hosts

Parasites with complex life cycles have the ability to change the behavior of their intermediate host in a way that increases their transmission rate to the next host. However, the level of behavioral changes can vary considerably, depending on the stage of parasite development and parasite intensity. To investigate the influence of such parameters, we evaluated the locomotory activity of the fish Poecilia vivipara prior to experimental infections, 7 days post-infection (dpi) and 14dpi with cercariae of the digenean Ascocotyle (Phagicola) pindoramensis. The locomotory activity was monitored using an image system, Videomex(®), linked to with a video camera able to record the swimming behavior of the fishes. At the end of the experiments, fishes were dissected and all metacercariae from the gills and mesenteries, the specific sites utilized by A. (P.) pindoramensis, were recovered and counted. There was a significant decrease in the swimming behavior of fishes after 14dpi. Similarly, we found a significant correlation between the swimming behavior of the fishes and parasite intensity in both sites of infection. It is surmised that the decrease in locomotory activity of P. vivipara caused by A. (P.) pindoramensis can disturb its predator-prey relationship in natural environment.

To Dash or to Dawdle: Verb-Associated Speed of Motion Influences Eye Movements during Spoken Sentence Comprehension

In describing motion events verbs of manner provide information about the speed of agents or objects in those events. We used eye tracking to investigate how inferences about this verb-associated speed of motion would influence the time course of attention to a visual scene that matched an event described in language. Eye movements were recorded as participants heard spoken sentences with verbs that implied a fast ("dash") or slow ("dawdle") movement of an agent towards a goal. These sentences were heard whilst participants concurrently looked at scenes depicting the agent and a path which led to the goal object. Our results indicate a mapping of events onto the visual scene consistent with participants mentally simulating the movement of the agent along the path towards the goal: when the verb implies a slow manner of motion, participants look more often and longer along the path to the goal; when the verb implies a fast manner of motion, participants tend to look earlier at the goal and less on the path. These results reveal that event comprehension in the presence of a visual world involves establishing and dynamically updating the locations of entities in response to linguistic descriptions of events.

Too risky to settle: avian community structure changes in response to perceived predation risk on adults and offspring

Predation risk is widely hypothesized as an important force structuring communities, but this potential force is rarely tested experimentally, particularly in terrestrial vertebrate communities. How animals respond to predation risk is generally considered predictable from species life-history and natural-history traits, but rigorous tests of these predictions remain scarce. We report on a large-scale playback experiment with a forest bird community that addresses two questions: (i) does perceived predation risk shape the richness and composition of a breeding bird community? And (ii) can species life-history and natural-history traits predict prey community responses to different types of predation risk? On 9 ha plots, we manipulated cues of three avian predators that preferentially prey on either adult birds or offspring, or both, throughout the breeding season. We found that increased perception of predation risk led to generally negative responses in the abundance, occurrence and/or detection probability of most prey species, which in turn reduced the species richness and shifted the composition of the breeding bird community. Species-level responses were largely predicted from the key natural-history trait of body size, but we did not find support for the life-history theory prediction of the relationship between species' slow/fast life-history strategy and their response to predation risk.

Assessing pneumococcal meningitis association with viral respiratory infections and antibiotics: insights from statistical and mathematical models

Pneumococcus is an important human pathogen, highly antibiotic resistant and a major cause of bacterial meningitis worldwide. Better prevention requires understanding the drivers of pneumococcal infection incidence and antibiotic susceptibility. Although respiratory viruses (including influenza) have been suggested to influence pneumococcal infections, the underlying mechanisms are still unknown, and viruses are rarely considered when studying pneumococcus epidemiology. Here, we propose a novel mathematical model to examine hypothetical relationships between Streptococcus pneumoniae meningitis incidence (SPMI), acute viral respiratory infections (AVRIs) and antibiotic exposure. French time series of SPMI, AVRI and penicillin consumption over 2001-2004 are analysed and used to assess four distinct virus-bacteria interaction submodels, ascribing the interaction on pneumococcus transmissibility and/or pathogenicity. The statistical analysis reveals strong associations between time series: SPMI increases shortly after AVRI incidence and decreases overall as the antibiotic-prescription rate rises. Model simulations require a combined impact of AVRI on both pneumococcal transmissibility (up to 1.3-fold increase at the population level) and pathogenicity (up to threefold increase) to reproduce the data accurately, along with diminished epidemic fitness of resistant pneumococcal strains causing meningitis (0.97 (0.96-0.97)). Overall, our findings suggest that AVRI and antibiotics strongly influence SPMI trends. Consequently, vaccination protecting against respiratory virus could have unexpected benefits to limit invasive pneumococcal infections.

Parental psychological distress during pregnancy and wheezing in preschool children: the Generation R Study

BACKGROUND

Maternal psychological distress during pregnancy might affect fetal lung development and subsequently predispose children to childhood asthma.

OBJECTIVE

We sought to assess the associations of maternal psychological distress during pregnancy with early childhood wheezing.

METHODS

We performed a population-based prospective cohort study among 4848 children. We assessed maternal and paternal psychological distress at the second trimester of gestation and 3 years after delivery and maternal psychological distress at 2 and 6 months after delivery by using the Brief Symptom Inventory questionnaire. Wheezing in the children was annually examined by using questionnaires from 1 to 4 years. Physician-diagnosed ever asthma was reported at 6 years.

RESULTS

Mothers with psychological distress during pregnancy had increased odds of wheezing in their children from 1 to 4 years of life (overall distress: odds ratio [OR], 1.60 [95% CI, 1.32-1.93]; depression: OR, 1.46 [95% CI, 1.20-1.77]; and anxiety: OR, 1.39 [95% CI, 1.15-1.67]). We observed similar positive associations with the number of wheezing episodes, wheezing patterns, and physician-diagnosed asthma at 6 years. Paternal distress during pregnancy and maternal and paternal distress after delivery did not affect these results and were not associated with childhood wheezing.

CONCLUSION

Maternal psychological distress during pregnancy is associated with increased odds of wheezing in their children during the first 6 years of life independent of paternal psychological distress during pregnancy and maternal and paternal psychological distress after delivery. These results suggest a possible intrauterine programming effect of maternal psychological distress leading to respiratory morbidity.

Reductions in HIV/STI incidence and sharing of injection equipment among female sex workers who inject drugs: results from a randomized controlled trial

BACKGROUND

We evaluated brief combination interventions to simultaneously reduce sexual and injection risks among female sex workers who inject drugs (FSW-IDUs) in Tijuana and Ciudad Juarez, Mexico during 2008-2010, when harm reduction coverage was expanding rapidly in Tijuana, but less so in Juarez.

METHODS

FSW-IDUs ≥18 years reporting sharing injection equipment and unprotected sex with clients within the last month participated in a randomized factorial trial comparing four brief, single-session conditions combining either an interactive or didactic version of a sexual risk intervention to promote safer sex in the context of drug use, and an injection risk intervention to reduce sharing of needles/injection paraphernalia. Women underwent quarterly interviews and testing for HIV, syphilis, gonorrhea, Chlamydia and Trichomonas, blinding interviewers and assessors to assignment. Poisson regression with robust variance estimation and repeated measures ordinal logistic regression examined effects on combined HIV/STI incidence and receptive needle sharing frequency.

FINDINGS

Of 584 initially HIV-negative FSW-IDUs, retention was ≥90%. After 12 months, HIV/STI incidence decreased >50% in the interactive vs. didactic sex intervention (Tijuana:AdjRR:0.38,95% CI:0.16-0.89; Juarez: AdjRR:0.44,95% CI:0.19-0.99). In Juarez, women receiving interactive vs. didactic injection risk interventions decreased receptive needle-sharing by 85% vs. 71%, respectively (p = 0.04); in Tijuana, receptive needle sharing declined by 95%, but was similar in active versus didactic groups. Tijuana women reported significant increases in access to syringes and condoms, but Juarez women did not.

INTERPRETATION

After 12 months in both cities, the interactive sexual risk intervention significantly reduced HIV/STI incidence. Expanding free access to sterile syringes coupled with brief, didactic education on safer injection was necessary and sufficient for achieving robust, sustained injection risk reductions in Tijuana. In the absence of expanding syringe access in Juarez, the injection risk intervention achieved significant, albeit more modest reductions, suggesting that community-level interventions incorporating harm reduction are more powerful than individual-level interventions.

TRIAL REGISTRATION

clinicaltrials.gov NCT00840658.