Intraclass correlation estimates in a school-based smoking prevention study. Outcome and mediating variables, by sex and ethnicity

Rank intraclass correlation for clustered data

Clustered data are common in biomedical research. Observations in the same cluster are often more similar to each other than to observations from other clusters. The intraclass correlation coefficient (ICC), first introduced by R. A. Fisher, is frequently used to measure this degree of similarity. However, the ICC is sensitive to extreme values and skewed distributions, and depends on the scale of the data. It is also not applicable to ordered categorical data. We define the rank ICC as a natural extension of Fisher's ICC to the rank scale, and describe its corresponding population parameter. The rank ICC is simply interpreted as the rank correlation between a random pair of observations from the same cluster. We also extend the definition when the underlying distribution has more than two hierarchies. We describe estimation and inference procedures, show the asymptotic properties of our estimator, conduct simulations to evaluate its performance, and illustrate our method in three real data examples with skewed data, count data, and three-level ordered categorical data.

National prevalence and risk factors for tungiasis in Kenya

BACKGROUND

Tungiasis is a highly neglected tropical skin disease caused by the sand flea, Tunga penetrans, the female of which burrows into the skin, causing pain and itching. The disease occurs throughout South America and sub-Saharan Africa but there are few systematic data on national disease burdens. The tungiasis research community is keen to develop survey methods to fill this gap. Here we used a school-based, thorough examination method to determine the prevalence and risk factors for tungiasis in Kenya.

METHODS

We conducted the first nationally representative survey of tungiasis, including nine counties covering the major ecological zones of Kenya. A stratified multistage random sampling was used to select 22 primary schools from each of the nine counties and to select up to 114 pupils aged 8 to 14 years in each school. Pupils were examined thoroughly for tungiasis. Two surveys were conducted, the first between May and July 2021 and the second between October 2021 and April 2023 when pupils were also interviewed for risk factors. Mixed effect logistic regression models were used to test associations of independent variables with tungiasis using the school as a random effect.

RESULTS

The overall prevalence of tungiasis in the first survey was 1.35% [95% confidence interval (CI): 1.15-1.59%], and 0.89% in the second survey. The prevalence ranged from 0.08% (95% CI: 0.01-0.59%) in Taita Taveta county to 3.24% (95% CI: 2.35-4.44%) in Kajiado county. Tungiasis infection was associated with county of residence, male sex [adjusted odds ratio (aOR) = 2.01, 95% CI: 1.52-2.67], and lower age (aOR = 0.81, 95% CI: 0.75-0.88). For the first time we demonstrate an association with attending public schools rather than private schools (aOR = 5.62, 95% CI: 1.20-26.22) and lower socioeconomic status (aOR = 0.10, 95% CI: 0.03-0.33). Using a rapid screening method of the top of feet only, would have missed 62.9% of all cases, 78.9% of mild cases and 20.0% of severe cases.

CONCLUSIONS

Tungiasis is widely but heterogeneously distributed across Kenya. School-based surveys offer an efficient strategy for mapping tungiasis distribution.

Intracluster correlation coefficients from school-based cluster randomized trials of interventions for improving health outcomes in pupils

BACKGROUND AND OBJECTIVES

To summarize intracluster correlation coefficient (ICC) estimates for pupil health outcomes from school-based cluster randomized trials (CRTs) across world regions and describe their relationship with study design characteristics and context.

METHODS

School-based CRTs reporting ICCs for pupil health outcomes were identified through a literature search of MEDLINE (via Ovid). ICC estimates were summarized both overall and for different categories of study characteristics.

RESULTS

Two hundred and forty-six articles reporting ICC estimates were identified. The median (interquartile range) ICC was 0.031 (0.011 to 0.08) at the school level (N = 210) and 0.063 (0.024 to 0.1) at the class level (N = 46). The distribution of ICCs at the school level was well described by the beta and exponential distributions. Besides larger ICCs in definitive trials than feasibility studies, there were no clear associations between study characteristics and ICC estimates.

CONCLUSION

The distribution of school-level ICCs worldwide was similar to previous summaries from studies in the United States. The description of the distribution of ICCs will help to inform sample size calculations and assess their sensitivity when designing future school-based CRTs of health interventions.

Intra-cluster correlations in socio-demographic variables and their implications: An analysis based on large-scale surveys in India

Individuals who share similar socio-economic and cultural characteristics also share similar health outcomes. Consequently, they have a propensity to cluster together, which results in positive intra-class correlation coefficients (ICCs) in their socio-demographic and behavioural characteristics. In this study, using data from four rounds of the National Family Health Survey (NFHS), we estimated the ICC for selected socio-demographic and behavioural characteristics in rural and urban areas of six states namely Assam, Gujarat, Kerala, Punjab, Uttar Pradesh, and West Bengal. The socio-demographic and behavioural characteristics included religion & caste of the household head, use of contraception & prevalence of anaemia among currently married women and coverage of full immunization services among children aged 12-23 months. ICC was computed at the level ofPrimary Sampling Units (PSUs), that is, villages in rural areas and census enumeration blocks in urban areas. Our research highlights high clustering in terms of religion and caste within PSUs in India. In NFHS-4, the ICCs for religion ranged from the lowest of 0.19 in rural areas of Kerala to the highest of 0.67 in urban areas of West Bengal. For the caste of the household head, the ICCs ranged from the lowest of 0.12 in the urban areas of Punjab to the highest of 0.46 in the rural areas of Assam. In most of the states selected for the study, the values of ICC were higher for the use of family planning methods than for full immunization. The value of ICC for use of contraception was highest for rural areas of Assam (0.15) followed by rural areas of Gujarat (0.13). A higher value of ICC has considerable implications for determining an effective sample size for large-scale surveys. Our findings agree with the fact that for a given cluster size, the higher the value of ICC, the higher is the loss in precision of the estimate. Knowing and taking into account ICCs can be extremely helpful in determining an effective sample size when designing a large-scale demographic and health survey to arrive at estimates of parameters with the desired precision.

Rationale, design, and methods for the development of a youth adapted Brief Tobacco Intervention plus automated text messaging for high school students

BACKGROUND

Tobacco use is increasing among youth in the U.S. The current study seeks to develop and pilot test a universal group-based Youth Brief Tobacco Intervention (Y-BTI) plus mobile phone automated text messaging (ATM) for 9th grade students to both prevent initiation among non-users and promote cessation among current users.

METHODS

This study will use a sequential, multi-method research design beginning with mixed methods formative work with 9th grade students to adapt the existing young adult Brief Tobacco Intervention (BTI) for youth and develop automated text messages. The formative work with students will identify salient themes and strategies for the interventions and optimal delivery schedules for the ATM intervention. The second phase of the study evaluates the Y-BTI and ATM through a pilot cluster randomized controlled trial that compares four treatment combinations: (1) Y-BTI + ATM, (2) Y-BTI alone, (3) ATM alone, or (4) standard of care. The Y-BTI is a single session, group-based intervention delivered in schools. The ATM intervention will provide comparable content to the Y-BTI but be delivered via text messaging 3-5 times per week for four weeks.

CONCLUSION

In order to curb the rise of tobacco use among youth, interventions that are easily implemented and disseminated need to be developed. We aim to build upon previous research by showing that a universal group-based Y-BTI and ATM are effective in reducing tobacco use among a 9th graders by preventing initiation among tobacco naïve youth and promoting cessation among current users.

An Internet Quiz Game Intervention for Adolescent Alcohol Drinking: A Clustered RCT

BACKGROUND AND OBJECTIVES

Interventions on adolescent drinking have yielded mixed results. We assessed the effectiveness of an Internet quiz game intervention compared to conventional health education.

METHODS

In this cluster randomized controlled trial with parallel group design, we randomly allocated 30 participating schools to the Internet quiz game intervention or the conventional health education (comparison) group, with 1:1 ratio. Students of Hong Kong secondary schools (aged 12-15 years) were recruited. The intervention was a 4-week Web-based quiz game competition in which participating students answered 1000 alcohol-related multiple-choice quiz questions. The comparison group received a printed promotional leaflet and hyperlinks to alcohol-related information.

RESULTS

Of 30 eligible schools, 15 (4294 students) were randomly assigned to the Internet quiz game intervention group and 15 (3498 students) to the comparison group. Average age of participants was 13.30 years. No significant between-group differences were identified at baseline. Overall retention rate for students was 86.0%. At 1-month follow-up, fewer students in the intervention group reported drinking (9.8% vs 12.1%, risk ratio 0.79, 95% confidence interval [CI] 0.68 to 0.92; P = .003), and those who drank reported drinking less alcohol (standardized difference β -0.06, 95% CI -0.11 to -0.01; P = .02). Between-group differences remained statistically significant at 3-month follow-up (10.4% vs 11.6%, risk ratio 0.86, 95% CI 0.74 to 0.999; P = .048; β -0.06, 95% CI -0.11 to -0.01; P = .02).

CONCLUSIONS

The Internet quiz game intervention reduced underage drinking by 21% at 1-month and 14% at 3-month follow-up compared with conventional health education.

Trends in and predictors of pregnancy termination among 15-24 year-old women in Nigeria: a multi-level analysis of demographic and health surveys 2003-2018

BACKGROUND

Three-quarters of pregnancy terminations in Africa are carried out in unsafe conditions. Unsafe abortion is the leading cause of maternal mortality among 15-24 year-old women in Sub-Saharan Africa. Greater understanding of the wider determinants of pregnancy termination in 15-24 year-olds could inform the design and development of interventions to mitigate the harm. Previous research has described the trends in and factors associated with termination of pregnancy for women of reproductive age in Nigeria. However, the wider determinants of pregnancy termination have not been ascertained, and data for all women have been aggregated which may obscure differences by age groups. Therefore, we examined the trends in and individual and contextual-level predictors of pregnancy termination among 15-24 year-old women in Nigeria.

METHODS

We analysed data from the 2003, 2008, 2013 and 2018 Nigerian Demographic and Health Surveys (NDHS) comprising 45,793 women aged 15-24 years. Trends in pregnancy termination across the four survey datasets were examined using bivariate analysis. Individual and contextual predictors of pregnancy termination were analysed using a three-level binary logistic regression analysis and are reported as adjusted odds ratios (aOR) with 95% confidence intervals (CI).

RESULTS

Trends in pregnancy termination declined from 5.8% in 2003 to 4.2% in 2013 then reversed to 4.9% in 2018. The declining trend was greater for 15-24 year-old women with higher socioeconomic status. Around 17% of the total variation in pregnancy termination was attributable to community factors, and 7% to state-level factors. Of all contextual variables considered, only contraceptive prevalence (proxy for reproductive health service access by young women) at community level was significant. Living in communities with higher contraceptive prevalence increased odds of termination compared with communities with lower contraceptive prevalence (aOR = 4.2; 95% CI 2.7-6.6). At the individual-level, sexual activity before age 15 increased odds of termination (aOR = 2.3; 95% CI 1.9-2.8) compared with women who initiated sexual activity at age 18 years or older, and married women had increased odds compared with never married women (aOR = 3.0; 95% CI 2.5-3.7).

CONCLUSION

Our findings highlight the importance of disaggregating data for women across the reproductive lifecourse, and indicates where tailored interventions could be targeted to address factors associated with pregnancy termination among young women in Nigeria.

Effects of the ACT OUT! Social Issue Theater Program on Social-Emotional Competence and Bullying in Youth and Adolescents: Protocol for a Cluster Randomized Controlled Trial

Background

Students in the United States spend a meaningful portion of their developmental lives in school. In recent years, researchers and educators have begun to focus explicitly on social and emotional learning (SEL) in the school setting. Initial evidence from meta-analyses suggests that curricula designed to promote SEL likely produce benefits in terms of social-emotional competence (SEC) and numerous related behavioral and affective outcomes. At the same time, there are often barriers to implementing such curricula as intended, and some researchers have questioned the strength of the evaluation data from SEL programs. As part of the effort to improve programming in SEL, this paper describes the protocol for a cluster randomized trial of the ACT OUT! Social Issue Theater program, a brief psychodramatic intervention to build SEC and reduce bullying behavior in students.

Objective

The objective of this trial is to examine if a short dose of interactive psychodrama can affect SEC metrics and bullying experiences in schoolchildren in either the short (2-week) or medium (6-month) term.

Methods

The ACT OUT! trial is a cluster randomized superiority trial with 2 parallel groups. The unit of measurement is the student, and the unit of randomization is the classroom. For each grade (fourth, seventh, and 10th), an even number of classrooms will be selected from each school—half will be assigned to the intervention arm and half will be assigned to the control arm. The intervention will consist of 3 moderated psychodramatic performances by trained actors, and the control condition will be the usual school day. Outcome data will be collected at baseline (preintervention), 2-week postintervention (short term), and 6-month postintervention (medium term). Outcomes will include social-emotional competency; self-reported bullying and experiences of being bullied; receptivity to the program; and school-level data on truancy, absenteeism, and referrals to school displinary action for bullying. A power analysis adjusted for clustering effect, design effect, and potential attrition yielded a need for approximately 1594 students, consisting of an estimated 80 classrooms split evenly into intervention and control arms.

Results

This study was funded in June 2019; approved by the Indiana University Institutional review board on September 17, 2019; began subject recruitment on November 5, 2019; and prospectively registered with ClinicalTrials.gov.

Conclusions

Many states have issued recommendations for the integration of SEL into schools. The proposed study uses a rigorous methodology to determine if the ACT OUT! psychodramatic intervention is a cost-effective means of bolstering SEC and reducing bullying incidence in schools.

Trial Registration

ClinicalTrials.gov NCT04097496; https://clinicaltrials.gov/ct2/show/NCT04097496

International Registered Report Identifier (IRRID)

PRR1-10.2196/17900

High risk for human exposure to Rift Valley fever virus in communities living along livestock movement routes: A cross-sectional survey in Kenya

Introduction

Multiple outbreaks of Rift Valley Fever (RVF) with devastating effects have occurred in East Africa. These outbreaks cause disease in both livestock and humans and affect poor households most severely. Communities living in areas practicing nomadic livestock movement may be at higher risk of infection. This study sought to i) determine the human exposure to Rift Valley fever virus (RVFV) in populations living within nomadic animal movement routes in Kenya; and ii) identify risk factors for RVFV infection in these communities.

Methods

A cross-sectional descriptive study design was used. Samples were collected from the year 2014 to 2015 in a community-based sampling exercise involving healthy individuals aged ≥18 years from Isiolo, Tana River, and Garissa counties. In total, 1210 samples were screened by ELISA for the presence of immunoglobulin IgM and IgG antibodies against RVFV. Positive results were confirmed by plaque reduction neutralization test.

Results

Overall, IgM and IgG prevalence for all sites combined was 1.4% (95% CI 0.8–2.3%) and 36.4% (95% CI 33.8–39.2%), respectively. Isiolo County recorded a non-significant higher IgG prevalence of 38.8% than Garissa 35.9% and Tana River 32.2% (Chi square = 2.5, df = 2, p = 0.287). Males were significantly at higher risk of infection by RVFV than females (OR = 1.67, 95% CI 1.17–2.39, p<0.005). Age was significantly associated with RVFV infection (Wald Chi = 94.2, df = 5, p<0.0001). Individuals who had regular contact with cattle (OR = 1.38, 95%CI 1.01–1.89) and donkeys (OR = 1.38, 95%CI 1.14–1.67), or contact with animals through birthing (OR = 1.69, 95%CI 1.14–2.51) were significantly at a greater risk of RVFV infection than those who did not.

Conclusion

This study demonstrated that although the Isiolo County has been classified as being at medium risk for RVF, virus infection appeared to be as prevalent in humans as in Tana River and Garissa, which have been classified as being at high risk. Populations in these counties live within nomadic livestock movement routes and therefore at risk of being exposed to the RVFV. Interventions to control RVFV infections therefore, should target communities living along livestock movement pathways.

Rift Valley fever (RVF) is a neglected mosquito-borne zoonotic disease that causes major outbreaks and economic harm to human and ruminants health leading to increased poverty within affected communities. RVF is caused by RVF virus (RVFV) affecting humans and a wide range of ruminants. The virus is transmitted through bites from mosquitoes and exposure to blood, body fluids, or tissues of infected ruminants. It was first isolated in Kenya in 1930 and several outbreaks have been recorded in many countries in sub-Saharan Africa. We studied pastoralist communities living along livestock migratory routes. Migratory livestock do move long distances in search of water and pasture and may be at higher risk of exposure to RVFV. We also determined risk factors for RVFV infection by studying age, gender, contact with animals through birthing, and occupation. Prevention and control of RVFV infection can target significant risk factors to prevent spread and re-occurrence of outbreaks.

Smoking-Cessation Interventions in Appalachia: A Systematic Review and Meta-Analysis

CONTEXT

Appalachia, a socioeconomically disadvantaged rural region in the eastern U.S., has one of the nation's highest prevalence rates of smoking and some of the poorest health outcomes. Effective interventions that lower smoking rates in Appalachia have great potential to reduce health disparities and preventable illness; however, a better understanding of effective interventions is needed.

EVIDENCE ACQUISITION

This review included trials that evaluated the impact of smoking-cessation programs among populations living in Appalachia. The search was carried out on October 9, 2018 and comprised the Cochrane Central Register of Controlled Trials, Medline, Embase, and Scopus for academic journal articles published in English, with no date restrictions. After preliminary screening, potentially relevant full-text articles were independently reviewed by the authors with a Cohen's κ of 0.72, leading to the final inclusion of 9 articles.

EVIDENCE SYNTHESIS

Eligible studies were assessed qualitatively for heterogeneity and risk of bias. Six of the 9 included studies had extractable data related to dichotomous smoking status and reported a measure of association suitable for inclusion in a meta-analysis. For those 6 studies, the pooled RR and pooled OR were estimated using random effects models, with an I² index demonstrating substantial heterogeneity. A funnel plot of the 6 trials appeared relatively symmetric.

CONCLUSIONS

Participation in smoking-cessation interventions increased the probability of smoking abstinence among Appalachian smokers by an estimated 2.33 times (pooled RR=2.33, 95% CI=1.03, 5.25, p=0.04). Given the low number of studies, their substantial heterogeneity, and high risk of bias, the evidence of the effectiveness of smoking-cessation interventions in Appalachia must be interpreted with caution.

Effects of the social norms intervention The GOOD Life on norm perceptions, binge drinking and alcohol-related harms: A cluster-randomised controlled trial

The study aimed to investigate if the school-based social norms intervention The GOOD Life was effective in reducing misperceptions, heavy alcohol use and alcohol-related harms among Danish pupils aged 13–17 years.

In total 38 schools were included in a cluster-randomised controlled trial and allocated to either intervention (n = 641) or control group (n = 714) during 2015/2016. Both groups completed an online survey before the intervention and 3 months after baseline. The GOOD Life intervention provided normative feedback tailored for each school-grade using three communication channels: classroom sessions, posters and web application. Outcome measures were overestimation of peers' lifetime binge drinking, binge drinking (5 or more drinks on one occasion) and alcohol-related harms. Intervention effects at follow-up were examined using multilevel logistic regression models.

Pupils in the intervention group were less likely to overestimate peers' lifetime binge drinking compared to those in the control group (OR: 0.52, 95%CI: 0.33–0.83) and were less likely to report two or more alcohol-related harms (OR: 0.59, 95%CI: 0.37–0.93). Overall, no significant effect of the intervention was found on binge drinking. However, among pupils stating it would be ok, if they drank more (n = 296), a preventive effect was found on binge drinking four or more times during the last 30 days (OR: 0.37, 95%CI: 0.15–0.95). Additionally, the intervention effect on overestimation was higher among pupils who reported binge drinking at baseline.

Receiving the intervention had a positive effect on norm perceptions and alcohol-related harms. We also found that the intervention effect differed by baseline status of alcohol use.

Assessing Construct Validity in Math Achievement: An Application of Multilevel Structural Equation Modeling (MSEM)

The purpose of the present study was to model math achievement at both the person and university levels of the analyses in order to understand the optimal factor structure of math competency. Data involved 2,881 students who took a national mathematics examination as part of their entry at the university public system in Saudi Arabia. Four factors from the National math examination comprised the math achievement measure, namely, numbers and operations, algebra and analysis, geometry and measurement, and, statistics and probabilities. Data were analyzed using the aggregate method and by use of Multilevel Structural Equation Modeling (MSEM). Results indicated that both a unidimensional and a 4-factor correlated model fitted the data equally well using aggregate data, where for reasons of parsimony the unidimensional model was the preferred choice with these data. When modeling data including clustering, results pointed to alternative factor structures at the person and university levels. Thus, a unidimensional model provided the best fit at the University level, whereas a four-factor correlated model was most descriptive for person level data. The optimal simple structure was evaluated using the Ryu and West (2009) methodology for partially saturating the MSEM model and also met criteria for discriminant validation as described in Gorsuch (1983). Furthermore, a university level variable, namely the year of establishment, pointed to the superiority of older institutions with regard to math achievement. It is concluded that ignoring a multilevel structure in the data may result in erroneous conclusions with regard to the optimal factor structure and the tests of structural models following that.

A hybrid design testing a 3-step implementation model for community scale-up of an HIV prevention intervention in rural Malawi: study protocol

BACKGROUND

Scaling-up evidence-based behavior change interventions can make a major contribution to meeting the UNAIDS goal of no new HIV infections by 2030. We developed an evidence-based peer group intervention for HIV prevention and testing in Malawi that is ready for wider dissemination. Our innovative approach turns over ownership of implementation to rural communities. We adapted a 3-Step Implementation Model (prepare, roll-out and sustain) for communities to use. Using a hybrid design, we simultaneously evaluate community implementation processes and program effectiveness.

METHODS

Three communities in southern Malawi begin implementation in randomly-assigned order using a stepped wedge design. Our evaluation sample size of 144 adults and 144 youth per community provides sufficient power to examine primary outcomes of condom use and HIV testing. Prior to any implementation, the first participants in all three communities are recruited and complete the Wave 1 baseline survey. Waves 2-4 surveys occur after each community completes roll-out. Each community follows the model's three steps. During Prepare, the community develops a plan and trains peer group leaders. During Roll-Out, peer leaders offer the program. During Sustain, the community makes and carries out plans to continue and expand the program and ultimately obtain local funding. We evaluate degree of implementation success (Aim 1) using the community's benchmark scores (e.g, # of peer groups held). We assess implementation process and factors related to success (Aim 2) using repeated interviews and observations, benchmarks from Aim 1 and fidelity assessments. We assess effectiveness of the peer group intervention when delivered by communities (Aim 3) using multi-level regression models to analyze data from repeated surveys. Finally, we use mixed methods analyses of all data to assess feasibility, acceptability and sustainability (Aim 4).

DISCUSSION

The project is underway, and thus far the first communities have enthusiastically begun implementation. We have had to make several modifications along the way, such as moving from rapid-tests of STIs to symptoms screening by a nurse due to problems with test reliability and availability. If successful, results will provide a replicable evidence-based model for future community implementation of this and other health interventions.

TRIAL REGISTRATION

Clinical Trials.gov NCT02765659 Registered May 6, 2016.

Behavioural incentive interventions for health behaviour change in young people (5-18 years old): A systematic review and meta-analysis

Physical inactivity, an unhealthy diet, smoking, and alcohol consumption are key determinants of morbidity and mortality. These health behaviours often begin at a young age and track into adulthood, emphasising a need for interventions in children and young people. Previous research has demonstrated the potential effectiveness of behavioural incentive (BI) interventions in adults. However, little is known about their effectiveness in children and adolescents. Eight bibliographic databases were searched. Eligibility criteria included controlled trials using behavioural incentives (rewards provided contingent on successful performance of the target behaviour) as an intervention component for health behaviour change in children and adolescents. Intervention effects (standardised mean differences or odds ratios) were calculated and pooled by health behaviour, using a random effects model. Twenty-two studies were included (of n = 8392 identified), 19 of which were eligible for meta-analysis: physical activity (n = 8); healthier eating (n = 3); and smoking (n = 8). There was strong evidence that behavioural incentives may encourage healthier eating behaviours, some evidence that behavioural incentives were effective for encouraging physical activity behaviour, and limited evidence to support the use of behavioural incentives for smoking cessation and prevention in adolescents. Findings suggest that behavioural incentives may encourage uptake and initiation of healthy eating and physical activity in young people. However, this is a limited evidence base and a wide range of incentive designs have yet to be explored. Future research should further investigate the acceptability of these intervention approaches for young people.

Impact of non-uniform correlation structure on sample size and power in multiple-period cluster randomised trials

Stepped wedge and cluster randomised crossover trials are examples of cluster randomised designs conducted over multiple time periods that are being used with increasing frequency in health research. Recent systematic reviews of both of these designs indicate that the within-cluster correlation is typically taken account of in the analysis of data using a random intercept mixed model, implying a constant correlation between any two individuals in the same cluster no matter how far apart in time they are measured: within-period and between-period intra-cluster correlations are assumed to be identical. Recently proposed extensions allow the within- and between-period intra-cluster correlations to differ, although these methods require that all between-period intra-cluster correlations are identical, which may not be appropriate in all situations. Motivated by a proposed intensive care cluster randomised trial, we propose an alternative correlation structure for repeated cross-sectional multiple-period cluster randomised trials in which the between-period intra-cluster correlation is allowed to decay depending on the distance between measurements. We present results for the variance of treatment effect estimators for varying amounts of decay, investigating the consequences of the variation in decay on sample size planning for stepped wedge, cluster crossover and multiple-period parallel-arm cluster randomised trials. We also investigate the impact of assuming constant between-period intra-cluster correlations instead of decaying between-period intra-cluster correlations. Our results indicate that in certain design configurations, including the one corresponding to the proposed trial, a correlation decay can have an important impact on variances of treatment effect estimators, and hence on sample size and power. An R Shiny app allows readers to interactively explore the impact of correlation decay.

Incentives for preventing smoking in children and adolescents

BACKGROUND

Adult smoking usually has its roots in adolescence. If individuals do not take up smoking during this period it is unlikely that they ever will. Further, once smoking becomes established, cessation is challenging; the probability of subsequently quitting is inversely proportional to the age of initiation. One novel approach to reducing the prevalence of youth smoking is the use of incentives.

OBJECTIVES

To assess the effect of incentives on preventing children and adolescents (aged 5 to 18 years) from starting to smoke. It was also our intention to assess, where possible, the dose-response of incentives, the costs of incentive programmes, whether incentives are more or less effective in combination with other interventions to prevent smoking initiation, and any unintended consequences arising from the use of incentives.

SEARCH METHODS

For the original review (published 2012) we searched the Cochrane Tobacco Addiction Group Specialized Register, with additional searches of MEDLINE, Embase, CINAHL, CSA databases and PsycINFO for terms relating to incentives, in combination with terms for smoking and tobacco use, and children and adolescents. The most recent searches were of the Cochrane Tobacco Addiction Group Specialized Register, and were carried out in December 2016.

SELECTION CRITERIA

We considered randomized controlled trials (RCTs) allocating children and adolescents (aged 5 to 18 years) as individuals, groups or communities to intervention or control conditions, where the intervention included an incentive aimed at preventing smoking uptake. We also considered controlled trials (CTs) with baseline measures and post-intervention outcomes.

DATA COLLECTION AND ANALYSIS

Two review authors extracted and independently assessed the data. The primary outcome was the smoking status of children or adolescents at follow-up who reported no smoking at baseline. We required a minimum follow-up of six months from baseline and assessed each included study for risks of bias. We used the most rigorous definition of abstinence in each trial; we did not require biochemical validation of self-reported tobacco use for study inclusion. Where possible we combined eligible studies to calculate pooled estimates at the longest follow-up, using the Mantel-Haenszel fixed-effect method, grouping studies by study design.

MAIN RESULTS

We identified three eligible RCTs and five CTs, including participants aged 11 to 14 years, who were non-smokers at baseline. Of the eight trials identified, six had analyzable data relevant for this review, which contributed to meta-analyses (7275 participants in total: 4003 intervention; 3272 control; 2484 participants after adjusting for clustering). All except one of the studies tested the 'Smokefree Class Competition' (SFC), which has been widely implemented throughout Europe. In this competition, classes with youth generally between the ages of 11 and 14 years commit to being smoke-free for a six-month period, and report their smoking status regularly. If 90% or more of the class are non-smokers at the end of the six months, the class goes into a competition to win prizes. The one study that was not a trial of the SFC was a controlled trial in which schools in two communities were assigned to the intervention, with schools in a third community acting as controls. Students in the intervention community with lower smoking rates at the end of the project (one school year) received rewards.Most studies resulted in statistically non-significant results. Only one study of the SFC reported a significant effect of the competition on the prevention of smoking at the longest follow-up. However, this study was at risk of multiple biases, and when we calculated the adjusted risk ratio (RR) we no longer detected a statistically significant difference. The pooled RR for the more robust RCTs (3 studies, n = 3056 participants/1107 adjusted for clustering) suggests that there is no statistically significant effect of incentives, in the form of the SFC, to prevent smoking initiation among children and adolescents in the long term (RR 1.00, 95% confidence interval (CI) 0.84 to 1.19). Pooled results from the non-randomized trials also did not detect a significant effect of the SFC, and we were unable to extract data on our outcome of interest from the one trial that did not study the SFC. There is little robust evidence to suggest that unintended consequences (such as making false claims about their smoking status and bullying of smoking students) are consistently associated with such interventions, although this has not been the focus of much research. There was insufficient information to assess the dose-response relationship or to report costs of incentives for preventing smoking uptake.We judged the included RCTs to be at unclear risk of bias, and the non-RCTs to be at high risk of bias. Using GRADE, we rated the overall quality of the evidence for our primary outcome as 'low' (for RCTs) and 'very low' (for non-RCTs), because of imprecision (all studies had wide confidence intervals), and for the risks of bias identified. We further downgraded the non-RCT evidence, due to issues with the non-RCT study design, likely to introduce further bias.

AUTHORS' CONCLUSIONS

The very limited evidence currently available suggests that incentive programmes do not prevent smoking initiation among youth. However, there are relatively few published studies and these are of variable quality. In addition, trials included in the meta-analyses were all studies of the SFC, which distributed small to moderately-sized prizes to whole classes, usually through a lottery system. It is therefore possible that other incentive programmes could be more successful at preventing smoking uptake in young people.Future studies might investigate the efficacy of a wider range of incentives, including those given to individual participants to prevent smoking uptake, whilst considering both the effect of incentives on smoking initiation and the progression to smoking. It would be useful if incentives were evaluated in varying populations from different socioeconomic and ethnic backgrounds, and if intervention components were described in detail.

Multilevel determinants of teenage childbearing in sub-Saharan Africa in the context of HIV/AIDS

This paper examined national variations and multilevel determinants of teenage childbearing in sub-Saharan Africa (SSA) in the context of HIV/AIDS using data from recent Demographic and Health Surveys conducted in 29 countries of SSA. Results showed significant community and national variations in teenage childbearing, partly explained by socio-economic and HIV/AIDS context. At community level, lower HIV/AIDS stigma, higher wealth and female education were associated with lower teenage childbearing. However, national socio-economic status had an intricate relationship with teenage childbearing. Higher national GDP per-capita was generally associated with higher teenage childbearing, and this relationship was stronger in lower HIV prevalence countries.

Impact of communities, health, and emotional-related factors on smoking use: comparison of joint modeling of mean and dispersion and Bayes' hierarchical models on add health survey

BACKGROUND

The analysis of correlated binary data is commonly addressed through the use of conditional models with random effects included in the systematic component as opposed to generalized estimating equations (GEE) models that addressed the random component. Since the joint distribution of the observations is usually unknown, the conditional distribution is a natural approach. Our objective was to compare the fit of different binary models for correlated data in Tabaco use. We advocate that the joint modeling of the mean and dispersion may be at times just as adequate. We assessed the ability of these models to account for the intraclass correlation. In so doing, we concentrated on fitting logistic regression models to address smoking behaviors.

METHODS

Frequentist and Bayes' hierarchical models were used to predict conditional probabilities, and the joint modeling (GLM and GAM) models were used to predict marginal probabilities. These models were fitted to National Longitudinal Study of Adolescent to Adult Health (Add Health) data for Tabaco use.

RESULTS

We found that people were less likely to smoke if they had higher income, high school or higher education and religious. Individuals were more likely to smoke if they had abused drug or alcohol, spent more time on TV and video games, and been arrested. Moreover, individuals who drank alcohol early in life were more likely to be a regular smoker. Children who experienced mistreatment from their parents were more likely to use Tabaco regularly.

CONCLUSIONS

The joint modeling of the mean and dispersion models offered a flexible and meaningful method of addressing the intraclass correlation. They do not require one to identify random effects nor distinguish from one level of the hierarchy to the other. Moreover, once one can identify the significant random effects, one can obtain similar results to the random coefficient models. We found that the set of marginal models accounting for extravariation through the additional dispersion submodel produced similar results with regards to inferences and predictions. Moreover, both marginal and conditional models demonstrated similar predictive power.

The GOOD life: Study protocol for a social norms intervention to reduce alcohol and other drug use among Danish adolescents

BACKGROUND

It is currently unknown if school-based social norms interventions are effective in preventing harmful alcohol consumption and other drug use among adolescents in Denmark. This paper describes the social norms-based programme The GOOD life and the design of a cluster-randomized controlled trial to test its effectiveness.

METHODS/DESIGN

The intervention The GOOD life is composed of three social norms components representing three different communication channels, namely face-to-face communication (normative feedback session), print communication (posters) and interactive media (web application). The intervention period of 8 weeks is preceded and followed by data collection, with the follow-up taking place 3 months after baseline. Public schools in the Region of Southern Denmark with grades 8 and 9 are invited to participate in the study and participating schools are randomly allocated to either intervention or control schools. The aim is to recruit a total of 39 schools and a sample of 1.400 pupils for the trial. An online questionnaire is conducted to examine the use of alcohol, tobacco and marijuana as well as the perceived frequency of use among peers of their own grade, which is measured before and after the intervention. Baseline data is used to develop social norms messages which are included in the three intervention components. Primary outcomes are binge drinking (more than 5 units at one occasion) and perceived frequency of binge drinking among peers, while smoking, marijuana use and alcohol-related harm will be assessed as secondary outcomes.

DISCUSSION

The GOOD life study will provide necessary insights on descriptive and injunctive norms regarding alcohol and other drug use among Danish adolescents. In addition, it will provide new knowledge and insight on the feasibility, implementation context and effectiveness of a newly developed social norms intervention in the Danish school context.

TRIAL REGISTRATION

Date of registration: 17 February 2016 (retrospectively registered) at Current Controlled Trials with study ID ISRCTN27491960.

Factors associated with married women's support of male circumcision for HIV prevention in Uganda: a population based cross-sectional study

BACKGROUND

Despite the protective effect of male circumcision (MC) against HIV in men, the acceptance of voluntary MC in priority countries for MC scale-up such as Uganda remains limited. This study examined the role of women's sociodemographic characteristics, knowledge of HIV and sexual bargaining power as determinants of women's support of male circumcision (MC).

METHODS

Data from the Uganda AIDS Indicator Survey, 2011 were analyzed (n = 4,874). Bivariate and multivariate logistic regression analyses with random intercept were conducted to identify factors that influence women's support of MC.

RESULTS

Overall, 67.0 % (n = 3,276) of the women in our sample were in support of MC but only 28.0 % had circumcised partners. Women who had the knowledge that circumcision reduces HIV risk were about 6 times as likely to support MC than women who lacked that knowledge [AOR (adjusted odds ratio) = 5.85, 95 % CI (confidence interval) = 4.83-7.10]. The two indicators of women's sexual bargaining power (i.e., ability to negotiate condom use and ability to refuse sex) were also positively associated with support of MC. Several sociodemographic factors particularly wealth index were also positively associated with women's support of MC.

CONCLUSIONS

The findings in this study will potentially inform intervention strategies to enhance uptake of male circumcision as a strategy to reduce HIV transmission in Uganda.

Estimates of Intraclass Correlation Coefficients and Other Design Parameters for Studies of School-Based Nutritional Interventions

INTRODUCTION

When cluster randomized trials are used to evaluate school-based nutritional interventions such as school lunch programs, design-stage estimates of the required sample size must take into account the correlation in outcomes among individuals within each cluster (e.g., classrooms, schools, or districts). Estimates of the necessary parameters have been carefully developed for educational interventions, but for nutritional interventions the literature is thin.

METHODS

Using data from two large multi-school, multi-district impact evaluations conducted in the United States, this article calculates estimates of the design parameters required for sizing school-based nutritional studies. The large size of the trials (252 and 1,327 schools) yields precise estimates of the parameters of interest. Variance components are estimated by fitting random-intercept multilevel models in Stata.

RESULTS

School-level intraclass correlations are similar to those typically found for educational outcomes. In particular, school-level estimates range from less than .01 to .26 across the two studies, and district-level estimates ranged from less than .01 to .19. This suggests that cluster randomized trials of nutritional interventions may require samples with numbers of schools similar to the education studies to detect similar effect sizes.

Predictors of Violent Behavior in an Early Adolescent Cohort: Similarities and Differences Across Genders

The authors assessed a cohort of 2,335 students from the Minneapolis, Minnesota, area to identify predictors of violent behavior and to determine whether the predictors varied by gender. The sample was 76% White; boys and girls were equally represented. The majority lived with two parents. A measure of violent behavior collected at the end of the eighth-grade year (2000) was entered into Poisson regression against baseline data collected at the beginning of the seventh-grade year (1998). Predictors of violent behavior influencing both boys and girls included depressive symptoms, perceived invulnerability to negative future events, paternal nonauthoritative behavior, and drinking alcohol. Additional predictors of violent behavior specific to girls included both risk and protective factors.

Methods for Multilevel Ordinal Data in Prevention Research

This paper discusses statistical models for multilevel ordinal data that may be more appropriate for prevention outcomes than models that assume continuous measurement and normality. Prevention outcomes often have distributions that make them inappropriate for many popular statistical models that assume normality and are more appropriately considered ordinal outcomes. Despite this, the modeling of ordinal outcomes is often not well understood. This article discusses ways to analyze multilevel ordinal outcomes that are clustered or longitudinal, including the proportional odds regression model for ordinal outcomes, which assumes that the covariate effects are the same across the levels of the ordinal outcome. The article will cover how to test this assumption and what to do if it is violated. It will also discuss application of these models using computer software programs.

An Internet-Based Intervention to Promote Alcohol-Related Attitudinal and Behavioral Change Among Adolescents: Protocol of a Cluster Randomized Controlled Trial

Background

Underage drinking is a prevalent risk behavior and common public health problem. Research shows that alcohol abuse not only affects the quality of life of drinkers themselves. The problems resulting from underage drinking pose substantial costs to society as well. The proposed study will address underage drinking with the use of an Internet campaign, which is a cost-effective way of tackling the problem.

Objective

The aims of this study are to test the effectiveness of an online quiz competition in changing adolescents’ alcohol-related attitudes and behavior and to explore the feasibility of using Internet viral marketing to reach a significant number of adolescents.

Methods

The study will constitute a cluster randomized controlled trial for 20 secondary schools (6720 Grade 7-9 students). Schools will be randomized to intervention or control arm with equal likelihood. Students in intervention schools will be invited to take part in the Internet campaign, whereas those in control schools will receive relevant promotional leaflets.

Results

Alcohol-related attitude and behavior will be the primary outcome measures. The results of the proposed study will provide evidence on the efficacy of an Internet intervention in modifying adolescents’ attitudes and behavior and guide further investigation into the prevention of and intervention in such risk behaviors as underage drinking. The project was funded July 2015, enrollment started September 2015, and results are expected July 2017.

Conclusions

With the Internet increasingly being recognized as a practical and cost-effective platform for health information delivery, the proposed Internet-based intervention is expected to be more effective in altering adolescents’ alcohol-related attitudes and behaviors than traditional health promotion.

ClinicalTrial

ClinicalTrials.gov NCT02450344; https://clinicaltrials.gov/ct2/show/NCT02450344 (Archived by WebCite at http://www.webcitation.org/6heB2zMBD)

Physician Advice to Adolescents About Smoking: Who Gets Advised and Who Benefits Most?

PURPOSE

The Clinical Practice Guidelines instruct physicians to ask their patients about smoking and to advise against tobacco use. Physicians are urged especially to attend to racial minorities and teens because of these groups' increased susceptibility to smoking. Research on race and physician advice against smoking has produced contradictory findings. The purpose of this study is to clarify the relationships between physician communication about tobacco, race, and smoking among adolescents.

METHODS

This cross-sectional retrospective study explored (1) racial differences in rates of receiving physician communication and (2) whether the relationship between physician communication and smoking among adolescents was moderated by race. Multiple measures of smoking status were used (e.g., intentions to quit, quit attempts, quits, relapse status). We used a large (N = 5,154), predominately African-American (82.9%) sample of 11th graders.

RESULTS

Regular smokers were more likely to be screened about smoking. African Americans were more frequently advised against tobacco than Caucasians. Among African Americans, nonsmokers were most likely to be both screened and advised; among Caucasians, regular were most likely to be screened and advised. Overall, physician intervention was associated with greater benefits for young African Americans, including fewer intentions to smoke, greater likelihood of quitting, and less relapse.

CONCLUSIONS

Physician communication about smoking may hold particular promise for African-American teens, reducing health disparities because of racial differences in smoking-related mortality and morbidity. Physicians should be encouraged to screen and advise all young people about tobacco, regardless of race or smoking status.

Estimates of Intraclass Correlation Coefficients From Longitudinal Group-Randomized Trials of Adolescent HIV/STI/Pregnancy Prevention Programs

INTRODUCTION

Group-randomized trials (GRTs) are one of the most rigorous methods for evaluating the effectiveness of group-based health risk prevention programs. Efficiently designing GRTs with a sample size that is sufficient for meeting the trial's power and precision goals while not wasting resources exceeding them requires estimates of the intraclass correlation coefficient (ICC)-the degree to which outcomes of individuals clustered within groups (e.g., schools) are correlated. ICC estimates vary widely depending on outcome, population, and setting, and small changes in ICCs can have large effects on the sample size needed to estimate intervention effects. This study addresses a gap in the literature by providing estimates of ICCs for adolescent sexual risk-taking outcomes under a range of study conditions.

METHOD

Multilevel regression analyses were applied to existing data from four federally funded GRTs of school-based HIV/STI/pregnancy prevention programs to obtain a variety of ICC estimates.

RESULTS

ICCs ranged from 0 to 0.15, with adjustment for covariates and repeated measurements reducing the ICC in the majority of cases. Minimum detectable effect sizes with 80% power and 0.05 significance levels ranged from small to medium Cohen's d (0.13 to 0.53) assuming 20 schools of 100 students each.

CONCLUSIONS

This study provides the first known set of ICC estimates for investigators to use when planning studies of school-based programs to prevent sexual risk behaviors in youth. The results provide further evidence of the importance of using the appropriate adjusted ICC estimate at the design stage to maximize resources in costly GRTs.

Methods for sample size determination in cluster randomized trials

BACKGROUND

The use of cluster randomized trials (CRTs) is increasing, along with the variety in their design and analysis. The simplest approach for their sample size calculation is to calculate the sample size assuming individual randomization and inflate this by a design effect to account for randomization by cluster. The assumptions of a simple design effect may not always be met; alternative or more complicated approaches are required.

METHODS

We summarise a wide range of sample size methods available for cluster randomized trials. For those familiar with sample size calculations for individually randomized trials but with less experience in the clustered case, this manuscript provides formulae for a wide range of scenarios with associated explanation and recommendations. For those with more experience, comprehensive summaries are provided that allow quick identification of methods for a given design, outcome and analysis method.

RESULTS

We present first those methods applicable to the simplest two-arm, parallel group, completely randomized design followed by methods that incorporate deviations from this design such as: variability in cluster sizes; attrition; non-compliance; or the inclusion of baseline covariates or repeated measures. The paper concludes with methods for alternative designs.

CONCLUSIONS

There is a large amount of methodology available for sample size calculations in CRTs. This paper gives the most comprehensive description of published methodology for sample size calculation and provides an important resource for those designing these trials.

Childhood bullying and social dilemmas

Children who witness bullying often do not defend victims. Bystanders might be reticent to intervene because they are stuck in "social dilemmas." Social dilemmas are situations in which individuals make decisions based on self-interest due to their lack of confidence that others will join with them in decisions that benefit the collective. In this study, the social dilemmas concept, which comes from game theory and social psychology, was applied to bullying for the first time. A total of 292 middle school students at a private residential school in the United States completed surveys about their bullying-related experiences within their residences of 10 to 12 students of the same gender. Multilevel modeling was employed to assess if and how attitudes, group norms, and social dilemmas predict behavior in bullying situations. The findings suggested that both individual and group factors were associated with behavior in bullying situations and that attitudes, group norms, and social dilemmas each made a unique contribution to predicting behavior in bullying situations. Aggr. Behav. 42:97-108, 2015. © 2015 Wiley Periodicals, Inc.

Effectiveness of a complex intervention in reducing the prevalence of smoking among adolescents: study design of a cluster-randomized controlled trial

Background

The likelihood of an adolescent taking up smoking may be influenced by his or her society, school and family. Thus, changes in the immediate environment may alter a young person’s perception of smoking.

Methods/Design

The proposed multi-center, cluster-randomized controlled trial will be stratified by the baseline prevalence of smoking in schools. Municipalities with fewer than 100,000 inhabitants will be randomly assigned to a control or intervention group. One secondary school will be randomly selected from each municipality. These schools will be randomized to two groups: the students of one will receive any existing educational course regarding smoking, while those of the other school will receive a four-year, class-based curriculum intervention (22 classroom lessons) aimed at reinforcing a smoke-free school policy and encouraging smoking cessation in parents, pupils, and teachers. The intervention will also include annual meetings with parents and efforts to empower adolescents to change the smoking-related attitudes and behaviors in their homes, classrooms and communities.

We will enroll children aged 12-13 years as they enter secondary school during two consecutive school years (to obtain sufficient enrolled subjects). We will follow them for five years, until two years after they leave secondary school. All external evaluators and analysts will be blinded to school allocation.

The aim of this study is to analyze the effectiveness of a complex intervention in reducing the prevalence of smoking in the third year of compulsory secondary education (ESO) and two years after secondary school, when the participants are 14-15 and 17-18 years old, respectively.

Discussion

Most interventions aimed at preventing smoking among adolescents yield little to no positive long-term effects. This clinical trial will analyze the effectiveness of a complex intervention aimed at reducing the incidence and prevalence of smoking in this vulnerable age group.

Trial registration

Current Controlled Trials: NCT01602796.

How stable are diurnal cortisol activity indices in healthy individuals? Evidence from three multi-wave studies

BACKGROUND

Indices of cortisol activity, including the cortisol awakening response (CAR), diurnal slope, and cortisol output across the day (total daily output), are often studied as mechanistic indicators that could link stress with health. Yet there is a paucity of data speaking to their temporal features, particularly whether they behave in a more state- or trait-like manner across time.

METHODS

To address this issue, data from 3 studies were used to assess CAR, diurnal slope and total daily output stability over different age groups and time spans: 130 healthy children and adolescents collected salivary cortisol samples 5 times/day (1, 4, 9 and 11h after wake) over 2 days at 5 visits spaced 6 months apart (Study 1); 147 adolescent girls collected saliva 6 times/day (wake, 1, 4, 9 and 14 h after wake) for 2 days at 3 visits, each a year apart (Study 2); and 47 healthy, primarily middle age adults collected saliva 6 times/day (wake, 1, 4, 9 and 14 h after wake) for 3 days at 4 visits spaced 2-3 months apart (Study 3). Stability was estimated by multilevel model-derived intraclass correlation coefficients (ICCs).

RESULTS

Across studies, approximately 50% of the variance in cortisol indices was attributable to day-to-day fluctuations, suggesting state-like properties. Of the indices, total daily output emerged as the most stable over time, followed by diurnal slope and CAR, but stability estimates were generally quite modest regardless of index and sample. Over time spans of >1 year, ICCs were ≤ .13.

CONCLUSIONS

Most of the variance in CAR, diurnal slope and total daily output reflects day-to-day fluctuation; there was little evidence for more stable trait-like influences. These findings suggest that future research should focus on short-term fluctuations in stress, cortisol and health, as opposed to lengthy disease processes.

Results of a multi-level intervention to prevent and control childhood obesity among Latino children: the Aventuras Para Niños Study

BACKGROUND

Community-based interventions are needed to reduce the burden of childhood obesity.

PURPOSE

To evaluate the impact of a multi-level promotora-based (Community Health Advisor) intervention to promote healthy eating and physical activity and prevent excess weight gain among Latino children.

METHODS

Thirteen elementary schools were randomized to one of four intervention conditions: individual/family level (Family-only), school/community level (Community-only), combined (Family + Community), or a measurement-only condition. Participants were 808 Latino parents and their children enrolled in kindergarten through 2(nd) grade. Measures included parent and child body mass index (BMI) and a self-administered parent survey that assessed several parent and child behaviors.

RESULTS

There were no significant intervention effects on children's BMI z-score. The family intervention changed several obesity-related child behaviors (e.g., fruit/vegetable consumption) and these were mediated by changes in parenting variables (e.g., parent monitoring).

CONCLUSION

A promotora-based behavioral intervention was efficacious at changing parental factors and child obesity-related health behaviors.

The disproportionate high risk of HIV infection among the urban poor in sub-Saharan Africa

The link between HIV infection and poverty in sub-Saharan Africa (SSA) is rather complex and findings from previous studies remain inconsistent. While some argue that poverty increases vulnerability, existing empirical evidence largely support the view that wealthier men and women have higher prevalence of HIV. In this paper, we examine the association between HIV infection and urban poverty in SSA, paying particular attention to differences in risk factors of HIV infection between the urban poor and non-poor. The study is based on secondary analysis of data from the Demographic and Health Surveys from 20 countries in SSA, conducted during 2003-2008. We apply multilevel logistic regression models, allowing the urban poverty risk factor to vary across countries to establish the extent to which the observed patterns are generalizable across countries in the SSA region. The results reveal that the urban poor in SSA have significantly higher odds of HIV infection than their urban non-poor counterparts, despite poverty being associated with a significantly lower risk among rural residents. Furthermore, the gender disparity in HIV infection (i.e. the disproportionate higher risk among women) is amplified among the urban poor. The paper confirms that the public health consequence of urban poverty that has been well documented in previous studies with respect to maternal and child health outcomes does apply to the risk of HIV infection. The positive association between household wealth and HIV prevalence observed in previous studies largely reflects the situation in the rural areas where the majority of the SSA populations reside.

Improving assessment of paediatric acute breathing difficulties in medical education: a cluster randomized controlled trial

AIM

Impact assessment of an adjunct to standard teaching on knowledge gain and self-assessed confidence for students undertaking their paediatric attachment and evaluation of an acute breathing difficulties (ABDs) learning package.

METHODS

Pragmatic cluster randomized controlled trial involving 248 UK medical undergraduates within six paediatric clinical rotations. Intervention groups received an online ABD learning package consisting of symptom-based decision-making pathways underpinned by a ratified ABD evidence-based guideline. Control groups received an online dummy package. Outcome measures were a pre- and post-intervention multiple choice question assessment and confidence questionnaire utilising 5-point Likert scales.

RESULTS

Significant knowledge gain in the intervention group (mean 6.84, 95% CI 5.56-8.12) versus no such difference in the control group, significant improvement in self-assessed confidence in intervention group versus no such difference in control group and significant confidence differences in ABD-specific statements were noted. In the satisfaction questionnaire, the learning content was noted to be new to only 16% participants.

CONCLUSION

Supplementation of standard teaching with an interactive delivery method resulted in cognitive gain and self-assessed confidence improvement in interaction with children with ABDs. This indicates that a well-designed tool can help prepare students for direct interaction with sick children as a junior clinician.

TRIAL REGISTRATION

ISRCTN27499282.

Incentives for preventing smoking in children and adolescents

BACKGROUND

Adult smoking usually has its roots in adolescence. If individuals do not take up smoking during this period it is unlikely that they ever will. Further, once smoking becomes established, cessation is challenging; the probability of subsequently quitting is inversely proportional to the age of initiation. One novel approach to reducing the prevalence of youth smoking is the use of incentives.

OBJECTIVES

To determine whether incentives prevent children and adolescents from starting to smoke. We also attempted to assess the dose-response of incentives, the costs of incentive programmes, whether incentives are more or less effective in combination with other interventions to prevent smoking initiation and any unintended consequences arising from the use of incentives.

SEARCH METHODS

We searched the Cochrane Tobacco Addiction Group Specialized Register, with additional searches of MEDLINE, EMBASE, CINAHL, CSA databases and PsycINFO for terms relating to incentives, in combination with terms for smoking and tobacco use, and children and adolescents. The most recent searches were in May 2012.

SELECTION CRITERIA

We considered randomized controlled trials allocating children and adolescents (aged 5 to 18 years) as individuals, groups or communities to intervention or control conditions, where the intervention included an incentive aimed at preventing smoking uptake. We also considered controlled trials with baseline measures and post-intervention outcomes.

DATA COLLECTION AND ANALYSIS

Data were extracted by two authors and assessed independently. The primary outcome was the smoking status of children or adolescents at follow-up who reported no smoking at baseline. We required a minimum follow-up of six months from baseline and assessed each included study for risk of bias. We used the most rigorous definition of abstinence in each trial; we did not require biochemical validation of self-reported tobacco use for study inclusion. Where possible we combined eligible studies to calculate pooled estimates at the longest follow-up using the Mantel-Haenszel fixed-effect method, grouping studies by study design.

MAIN RESULTS

We identified seven controlled studies that met our inclusion criteria, including participants with an age range of 11 to 14 years. Of the seven trials identified, only five had analysable data relevant for this review and contributed to the meta-analysis (6362 participants in total who were non-smokers at baseline; 3466 in intervention and 2896 in control). All bar one of the studies was a trial of the so-called Smokefree Class Competition (SFC), which has been widely implemented throughout Europe. In this competition, classes with youth generally between the ages of 11 to 14 years commit to being smoke free for a six month period. They report regularly on their smoking status; if 90% or more of the class is non-smoking at the end of the six months, the class goes into a competition to win prizes. The one study that was not a trial of the SFC was a controlled trial in which schools in two communities were assigned to the intervention, with schools in a third community acting as controls. Students in the intervention community with lower smoking rates at the end of the project (one school year) received rewards.Only one study of the SFC competition, a non-randomized controlled trial, reported a significant effect of the competition on the prevention of smoking at the longest follow-up. However, this study had a risk of multiple biases, and when we calculated the adjusted RR we no longer detected a statistically significant difference. The pooled RR for the more robust RCTs (3 studies, n = 3056 participants) suggests that, from the available data, there is no statistically significant effect of incentives to prevent smoking initiation among children and adolescents in the long term (RR 1.00, 95% CI 0.84 to 1.19). Pooled results from non-randomized trials also did not detect a significant effect, and we were unable to extract data on our outcome of interest for the one trial that did not study the SFC. There is little robust evidence to suggest that unintended consequences (such as youth making false claims about their smoking status and bullying of smoking students) are consistently associated with such interventions, although this has not been the focus of much research. There was insufficient information to assess the dose-response relationship or to report costs.

AUTHORS' CONCLUSIONS

To date, incentive programmes have not been shown to prevent smoking initiation among youth, although there are relatively few published studies and these are of variable quality. Trials included in this meta-analysis were all studies of the SFC competition, which distributed small to moderately sized prizes to whole classes, usually through a lottery system.Future studies might investigate the efficacy of incentives given to individual participants to prevent smoking uptake. Future research should consider the efficacy of incentives on smoking initiation, as well as progression of smoking, evaluate these in varying populations from different socioeconomic and ethnic backgrounds, and describe the intervention components in detail.

Multilevel factorial experiments for developing behavioral interventions: power, sample size, and resource considerations

Factorial experimental designs have many potential advantages for behavioral scientists. For example, such designs may be useful in building more potent interventions by helping investigators to screen several candidate intervention components simultaneously and to decide which are likely to offer greater benefit before evaluating the intervention as a whole. However, sample size and power considerations may challenge investigators attempting to apply such designs, especially when the population of interest is multilevel (e.g., when students are nested within schools, or when employees are nested within organizations). In this article, we examine the feasibility of factorial experimental designs with multiple factors in a multilevel, clustered setting (i.e., of multilevel, multifactor experiments). We conduct Monte Carlo simulations to demonstrate how design elements-such as the number of clusters, the number of lower-level units, and the intraclass correlation-affect power. Our results suggest that multilevel, multifactor experiments are feasible for factor-screening purposes because of the economical properties of complete and fractional factorial experimental designs. We also discuss resources for sample size planning and power estimation for multilevel factorial experiments. These results are discussed from a resource management perspective, in which the goal is to choose a design that maximizes the scientific benefit using the resources available for an investigation.

Community epidemiology of risk and adolescent substance use: practical questions for enhancing prevention

To promote an effective approach to prevention, the community diagnosis model helps communities systematically assess and prioritize risk factors to guide the selection of preventive interventions. This increasingly widely used model relies primarily on individual-level research that links risk and protective factors to substance use outcomes. I discuss common assumptions in the translation of such research concerning the definition of risk factor elevation; the equivalence, independence, and stability of relations between risk factors and problem behaviors; and community differences in risk factors and risk factor-problem behavior relations. Exploring these assumptions could improve understanding of the relations of risk factors and substance use within and across communities and enhance the efficacy of the community diagnosis model. This approach can also be applied to other areas of public health where individual and community levels of risk and outcomes intersect.

Sample Size Issues for Cluster Randomized Trials With Discrete-Time Survival Endpoints

With cluster randomized trials complete groups of subjects are randomized to treatment conditions. An important question might be whether and when the subjects experience a particular event, such as smoking initiation or recovery from disease. In the social sciences the timing of such events is often measured in discrete time by using time intervals. At the planning phase of a cluster randomized trial one should decide on the number of clusters and cluster size such that parameters are estimated accurately and sufficient power on the test on treatment effect is achieved. On basis of a simulation study it is concluded that regression coefficients are estimated more accurately than the variance of the random cluster effect. In addition, it is shown that power increases with cluster size and number of clusters, and that a sufficient power cannot always be achieved by using larger cluster sizes at a fixed number of clusters.

Reaching Jewish ultra-orthodox adolescents: results from a targeted smoking prevention trial

BACKGROUND

Ultra-orthodox, Jewish adolescent boys are considered to have relatively high smoking rates, but are generally not targeted by Israel's smoking prevention programs.

OBJECTIVE

The objective of this trial was to test the effectiveness of a religion-based tobacco control intervention in reducing smoking prevalence among these youth.

METHODS

The study population participants were 340 boys from 63 religious boys' schools in Jerusalem. The intervention consisted of a mailing that included a pamphlet describing the health effects of and rabbinical prohibitions on smoking. A cluster randomized trial was conducted between March and May, 2005. The primary endpoint was current smoking status. Secondary endpoints were future intent to smoke and attitudes towards smoking. Generalized estimating equations and mixed models of analysis of variance were used to perform the analyses.

RESULTS

The intervention did not significantly affect current smoking, intent to smoke or attitudes towards smoking. Prevalence of smoking and future intent to smoke were higher in schools without enforced smoking regulations [odds ratio (OR) 2.74, p=0.026, OR 3.38, p=0.018]. Increased smoking prevalence was associated with a high prevalence of smoking among friends (p=0.031) and not finding smoking repulsive (p=0.024).

CONCLUSIONS

This study adds to the public health literature linking smoke-free schools and peer influences to adolescent smoking. Pamphlets containing rabbinic prohibitions on smoking initiation did not affect smoking behavior or intent to smoke.

Reduction in overweight and obesity from a 3-year community-based intervention in Australia: the 'It's Your Move!' project

'It's Your Move!' was a 3-year intervention study implemented in secondary schools in Australia as part of the Pacific Obesity Prevention In Communities Project. This paper reports the outcome results of anthropometric indices and relevant obesity-related behaviours. The interventions focused on building the capacity of families, schools and communities to promote healthy eating and physical activity. Baseline response rates and follow-up rates were 53% and 69% respectively for the intervention group (n=5 schools) and 47% and 66% respectively for the comparison group (n=7 schools). Statistically significant relative reductions in the intervention versus comparison group were observed: weight (-0.74 kg, P < 0.04), and standardized body mass index (-0.07, P<0.03), and non-significant reductions in prevalence of overweight and obesity (0.75 odds ratio, P=0.12) and body mass index (-0.22, P=0.06). Obesity-related behavioural variables showed mixed results with no pattern of positive intervention outcomes. In conclusion, this is the first study to show that long-term, community-based interventions using a capacity-building approach can prevent unhealthy weight gain in adolescents. Obesity prevention efforts in this important transitional stage of life can be successful and these findings need to be translated to scale for a national effort to reverse the epidemic in children and adolescents.

Effects of physical activity on teen smoking cessation

OBJECTIVE

To understand the influence of physical activity on teen smoking-cessation outcomes.

METHODS

Teens (N = 233; 14-19 years of age) from West Virginia high schools who smoked >1 cigarette in the previous 30 days were included. High schools with >300 students were selected randomly and assigned to brief intervention (BI), Not on Tobacco (N-O-T) (a proven teen cessation program), or N-O-T plus a physical activity module (N-O-T+FIT). Quit rates were determined 3 and 6 months after baseline by using self-classified and 7-day point prevalence quit rates, and carbon monoxide validation was obtained at the 3-month follow-up evaluation.

RESULTS

Trends for observed and imputed self-classified and 7-day point prevalence rates indicated that teens in the N-O-T+FIT group had significantly higher cessation rates compared with those in the N-O-T and BI groups. Effect sizes were large. Overall, girls quit more successfully with N-O-T compared with BI (relative risk [RR]: >∞) 3 months after baseline, and boys responded better to N-O-T+FIT than to BI (RR: 2-3) or to N-O-T (RR: 1-2). Youths in the N-O-T+FIT group, compared with those in the N-O-T group, had greater likelihood of cessation (RR: 1.48) at 6 months. The control group included an unusually large proportion of participants in the precontemplation stage at enrollment, but there were no significant differences in outcomes between BI and N-O-T (z = 0.94; P = .17) or N-O-T+FIT (z = 1.12; P = .13) participants in the precontemplation stage.

CONCLUSIONS

Adding physical activity to N-O-T may enhance cessation success, particularly among boys.

A school-based resilience intervention to decrease tobacco, alcohol and marijuana use in high school students

Background

Despite schools theoretically being an ideal setting for accessing adolescents and preventing initiation of substance use, there is limited evidence of effective interventions in this setting. Resilience theory provides one approach to achieving such an outcome through improving adolescent mental well-being and resilience. A study was undertaken to examine the potential effectiveness of such an intervention approach in improving adolescent resilience and protective factor scores; and reducing the prevalence of adolescent tobacco, alcohol and marijuana use in three high schools.

Methods

A non-controlled before and after study was undertaken. Data regarding student resilience and protective factors, and measures of tobacco, alcohol and marijuana use were collected from grade 7 to 10 students at baseline (n = 1449) and one year following a three year intervention (n = 1205).

Results

Significantly higher resilience and protective factors scores, and significantly lower prevalence of substance use were evident at follow up.

Conclusions

The results suggest that the intervention has the potential to increase resilience and protective factors, and to decrease the use of tobacco, alcohol and marijuana by adolescents. Further more rigorous research is required to confirm this potential.

A multilevel analysis of the determinants and cross-national variations of HIV seropositivity in sub-Saharan Africa: evidence from the DHS

This paper applies multilevel logistic regression models to Demographic and Health Survey data collected during 2003–2008 from 20 countries of sub-Saharan Africa to examine the determinants and cross-national variations in the risk of HIV seropositivity in the region. The models include individual-level and contextual region/country-level risk factors. Simultaneous confidence intervals of country-level residuals are used to compare the risk of being HIV seropositive across countries. The study reveals interesting general patterns in the risk of HIV seropositivity in sub-Saharan Africa. In particular, the findings highlight the gender disparity in socio-economic risk factors, partly explained by sexual behaviour factors.

Household and community HIV/AIDS status and child malnutrition in sub-Saharan Africa: evidence from the demographic and health surveys

This paper examines the extent to which under five children in households or communities adversely affected by HIV/AIDS are disadvantaged, in comparison with other children in less affected households/communities. The study is based on secondary analysis of the Demographic and Health Survey (DHS) data collected during 2003-2008 from 18 countries in sub-Saharan Africa, where the DHS has included HIV test data for adults of reproductive age. We apply multilevel logistic regression models that take into account the effect of contextual community/country level HIV/AIDS factors on child malnutrition. The outcome variable of interest is child undernutrition: stunting, wasting and underweight. The results suggest that across countries in sub-Saharan Africa, children whose mothers are infected with HIV are significantly more likely to be stunted, wasted or underweight compared to their counterparts of similar demographic and socio-economic background whose mothers are not infected. However, the nutritional status of children who are paternal orphans or in households where other adults are HIV positive are not significantly different from non-orphaned children or those in households where no adult is infected with HIV. Other adult household members being HIV positive is, however, associated with higher malnutrition among younger children below the age of one. Further analysis reveals that the effect of mothers' HIV status on child nutritional status (underweight) varies significantly across communities within countries, the effect being lower in communities with generally higher levels of malnutrition. Overall, the findings have important implications for policy and programme efforts towards improved integration of HIV/AIDS and child nutrition services in affected communities and other sub-groups of the population made vulnerable by HIV/AIDS. In particular, children whose mothers are infected with HIV deserve special attention.

Cross-national analysis of the risk factors of child malnutrition among children made vulnerable by HIV/AIDS in sub-Saharan Africa: evidence from the DHS

Objective

To examine the risk factors of malnutrition among children whose mothers are infected with HIV in sub-Saharan Africa (SSA).

Methods

Multilevel logistic regression models applied to Demographic and Health Survey (DHS) data collected during 2003–2008 from 18 countries in SSA, where the DHS included HIV test data for adults of reproductive age.

Results

Across countries in SSA, the risk of malnutrition among children whose mothers are infected with HIV is particularly high among children aged one, boys, multiple/twin births, those who were smaller than average at birth, or whose mothers had no education, or in poorest or single parent households. Although these risk factors generally apply to all children from the same communities, the higher risk of child malnutrition among those in the poorest households is amplified among children whose mothers are infected with HIV. Also, while in general children who are breastfed for up to 6 months are significantly less likely to be malnourished than those who were never breastfed; the benefit of breastfeeding is not evident among children whose mothers are infected with HIV.

Conclusion

Contextual community/country HIV prevalences show interesting patterns: the risk of malnutrition among children whose mothers are infected with HIV is lower in countries with higher HIV prevalence. These findings have important implications for interventions to address malnutrition among children made vulnerable by HIV/AIDS in the SSA region.