Comparison of methods for estimating the intraclass correlation coefficient for binary responses in cancer prevention cluster randomized trials

Wu, Shenghao; Crespi, Catherine M.; Wong, Weng Kee

doi:10.1016/j.cct.2012.05.004

Cited by 208 publications

(179 citation statements)

References 39 publications

Supporting

Mentioning

171

Contrasting

Unclassified

Order By: Relevance

“…29 Similar results were observed for 5-to 16-year-old children who had been prescribed preventative steroids. Among children aged under 5 years, the differences were larger, and of borderline statistical significance, with the intervention being associated with more unscheduled visits for all subgroups.…”

Section: Unscheduled Medical Contacts In September 2013supporting

confidence: 75%

PLEASANT: Preventing and Lessening Exacerbations of Asthma in School-age children Associated with a New Term – a cluster randomised controlled trial and economic evaluation

Julious

Horspool

Davis

et al. 2016

Health Technol Assess

View full text Add to dashboard Cite

BackgroundAsthma episodes and deaths are known to be seasonal. A number of reports have shown peaks in asthma episodes in school-aged children associated with the return to school following the summer vacation. A fall in prescription collection in the month of August has been observed, and was associated with an increase in the number of unscheduled contacts after the return to school in September.ObjectiveThe primary objective of the study was to assess whether or not a NHS-delivered public health intervention reduces the September peak in unscheduled medical contacts.DesignCluster randomised trial, with the unit of randomisation being 142 NHS general practices, and trial-based economic evaluation.SettingPrimary care.InterventionA letter sent (n = 70 practices) in July from their general practitioner (GP) to parents/carers of school-aged children with asthma to remind them of the importance of taking their medication, and to ensure that they have sufficient medication prior to the start of the new school year in September. The control group received usual care.Main outcome measuresThe primary outcome measure was the proportion of children aged 5–16 years who had an unscheduled medical contact in September 2013. Supporting end points included the proportion of children who collected prescriptions in August 2013 and unscheduled contacts through the following 12 months. Economic end points were quality-adjusted life-years (QALYs) gained and costs from an NHS and Personal Social Services perspective.ResultsThere is no evidence of effect in terms of unscheduled contacts in September. Among children aged 5–16 years, the odds ratio (OR) was 1.09 [95% confidence interval (CI) 0.96 to 1.25] against the intervention. The intervention did increase the proportion of children collecting a prescription in August (OR 1.43, 95% CI 1.24 to 1.64) as well as scheduled contacts in the same month (OR 1.13, 95% CI 0.84 to 1.52). For the wider time intervals (September–December 2013 and September–August 2014), there is weak evidence of the intervention reducing unscheduled contacts. The intervention did not reduce unscheduled care in September, although it succeeded in increasing the proportion of children collecting prescriptions in August as well as having scheduled contacts in the same month. These unscheduled contacts in September could be a result of the intervention, as GPs may have wanted to see patients before issuing a prescription. The economic analysis estimated a high probability that the intervention was cost-saving, for baseline-adjusted costs, across both base-case and sensitivity analyses. There was no increase in QALYs.LimitationThe use of routine data led to uncertainty in the coding of medical contacts. The uncertainty was mitigated by advice from a GP adjudication panel.ConclusionsThe intervention did not reduce unscheduled care in September, although it succeeded in increasing the proportion of children both collecting prescriptions and having scheduled contacts in August. After September there is weak evidence in favour of the intervention. The intervention had a favourable impact on costs but did not demonstrate any impact on QALYs. The results of the trial indicate that further work is required on assessing and understanding adherence, both in terms of using routine data to make quantitative assessments, and through additional qualitative interviews with key stakeholders such as practice nurses, GPs and a wider group of children with asthma.Trial registrationCurrent Controlled Trials ISRCTN03000938.Funding detailsThis project was funded by the NIHR Health Technology Assessment programme and will be published in full inHealth Technology Assessment; Vol. 20, No. 93. See the HTA programme website for further project information.

show abstract

Section: Unscheduled Medical Contacts In September 2013supporting

confidence: 75%

PLEASANT: Preventing and Lessening Exacerbations of Asthma in School-age children Associated with a New Term – a cluster randomised controlled trial and economic evaluation

Julious

Horspool

Davis

et al. 2016

Health Technol Assess

View full text Add to dashboard Cite

show abstract

“…This indicates that 28% of the chances of owning an album is explained by between-classroom differences (and -conversely -that 72% is explained by withinclassroom differences). For more detailed information on intraclass correlation coefficient in multilevel logistic regression, see Wu, Crespi, and Wong (2012).…”

Section: Step #1 Building An Empty Model: To What Extent Do the Log-mentioning

confidence: 99%

Keep Calm and Learn Multilevel Logistic Modeling: A Simplified Three-Step Procedure Using Stata, R, Mplus, and SPSS

Sommet¹,

Morselli

2017

International Review of Social Psychology

432

296

View full text Add to dashboard Cite

This paper aims to introduce multilevel logistic regression analysis in a simple and practical way. First, we introduce the basic principles of logistic regression analysis (conditional probability, logit transformation, odds ratio). Second, we discuss the two fundamental implications of running this kind of analysis with a nested data structure: In multilevel logistic regression, the odds that the outcome variable equals one (rather than zero) may vary from one cluster to another (i.e. the intercept may vary) and the effect of a lower-level variable may also vary from one cluster to another (i.e. the slope may vary). Third and finally, we provide a simplified three-step "turnkey" procedure for multilevel logistic regression modeling:• Preliminary phase: Cluster-or grand-mean centering variables • Step #1: Running an empty model and calculating the intraclass correlation coefficient (ICC) • Step #2: Running a constrained and an augmented intermediate model and performing a likelihood ratio test to determine whether considering the cluster-based variation of the effect of the lowerlevel variable improves the model fit • Step #3 Running a final model and interpreting the odds ratio and confidence intervals to determine whether data support your hypothesisCommand syntax for Stata, R, Mplus, and SPSS are included. These steps will be applied to a study on Justin Bieber, because everybody likes Justin Bieber.1

show abstract

“…There are various methods available to calculate ICC for a binary outcome, as discussed by Wu et al (2012) for randomized controlled trials and Fenn et al (2004) for crosssectional survey data. The ICC estimation from the random intercept logistic method suggested by Wu et al has some limitations.…”

Section: Methodsmentioning

confidence: 99%

“…This model partitions the variance in the outcome variable into 2 independent components: σ The sampling distribution of the variance estimates in multilevel logistic regression models is, in general, strongly asymmetric. Therefore, the standard error (SE) may be a poor characterization of the distribution, and confidence intervals (CIs) derived from the SE are likely to be unrepresentative of the data (Wu et al 2012). Given this difficulty, we estimated the 95% CI of the ICC and DE by using "bootstrapping"-a technique for generating a description of the sampling properties of empirical estimators using random sampling with replacement from the original data set (Hox 2010).…”

Section: Methodsmentioning

confidence: 99%

The Clustering Effects of Surfaces within the Tooth and Teeth within Individuals

Masood

Newton

2014

J Dent Res

View full text Add to dashboard Cite

The objectives of this study were 1) to provide an estimate of the value of the intraclass correlation coefficient (ICC) for dental caries data at tooth and surface level, 2) to provide an estimate of the design effect (DE) to be used in the determination of sample size estimates for future dental surveys, and 3) to explore the usefulness of multilevel modeling of cross-sectional survey data by comparing the model estimates derived from multilevel and single-level models. Using data from the United Kingdom Adult Dental Health Survey 2009, the ICC and DE were calculated for surfaces within a tooth, teeth within the individual, and surfaces within the individual. Simple and multilevel logistic regression analysis was performed with the outcome variables carious tooth or surface. ICC estimated that 10% of the variance in surface caries is attributable to the individual level and 30% of the variance in surfaces caries is attributable to variation between teeth within individuals. When comparing multilevel with simple logistic models, β values were 4 to 5 times lower and the standard error 2 to 3 times lower in multilevel models. All the fit indices showed multilevel models were a better fit than simple models. The DE was 1.4 for the clustering of carious surfaces within teeth, 6.0 for carious teeth within an individual, and 38.0 for carious surfaces within the individual. The ICC for dental caries data was 0.21 (95% confidence interval [CI], 0.204-0.220) at the tooth level and 0.30 (95% CI, 0.284-0.305) at the surface level. The DE used for sample size calculation for future dental surveys will vary on the level of clustering, which is important in the analysis-the DE is greatest when exploring the clustering of surfaces within individuals. Failure to consider the effect of clustering on the design and analysis of epidemiological trials leads to an overestimation of the impact of interventions and the importance of risk factors in predicting caries outcome.

show abstract

Comparison of methods for estimating the intraclass correlation coefficient for binary responses in cancer prevention cluster randomized trials

Cited by 208 publications

References 39 publications

PLEASANT: Preventing and Lessening Exacerbations of Asthma in School-age children Associated with a New Term – a cluster randomised controlled trial and economic evaluation

PLEASANT: Preventing and Lessening Exacerbations of Asthma in School-age children Associated with a New Term – a cluster randomised controlled trial and economic evaluation

Keep Calm and Learn Multilevel Logistic Modeling: A Simplified Three-Step Procedure Using Stata, R, Mplus, and SPSS

The Clustering Effects of Surfaces within the Tooth and Teeth within Individuals

Contact Info

Product

Resources

About