Review of methods for handling confounding by cluster and informative cluster size in clustered data

Seaman, Shaun; Pavlou, Menelaos; Copas, Andrew

doi:10.1002/sim.6277

Cited by 76 publications

(114 citation statements)

References 64 publications

Supporting

Mentioning

114

Contrasting

Order By: Relevance

“…The problem of arm sizes is closely bound up with that of missing data; for example, we have discussed how bias would arise if a study arm were included or excluded dependent on the results expected in that study arm, but bias would also arise if a study arm's size was chosen dependent on the results expected in that study arm. Methods for informative cluster size may be useful here.…”

Section: Discussionmentioning

confidence: 99%

A comparison of arm‐based and contrast‐based models for network meta‐analysis

White

Turner

Karahalios

et al. 2019

Statistics in Medicine

View full text Add to dashboard Cite

Differences between arm‐based (AB) and contrast‐based (CB) models for network meta‐analysis (NMA) are controversial. We compare the CB model of Lu and Ades (2006), the AB model of Hong et al(2016), and two intermediate models, using hypothetical data and a selected real data set. Differences between models arise primarily from study intercepts being fixed effects in the Lu‐Ades model but random effects in the Hong model, and we identify four key difference. (1) If study intercepts are fixed effects then only within‐study information is used, but if they are random effects then between‐study information is also used and can cause important bias. (2) Models with random study intercepts are suitable for deriving a wider range of estimands, eg, the marginal risk difference, when underlying risk is derived from the NMA data; but underlying risk is usually best derived from external data, and then models with fixed intercepts are equally good. (3) The Hong model allows treatment effects to be related to study intercepts, but the Lu‐Ades model does not. (4) The Hong model is valid under a more relaxed missing data assumption, that arms (rather than contrasts) are missing at random, but this does not appear to reduce bias. We also describe an AB model with fixed study intercepts and a CB model with random study intercepts. We conclude that both AB and CB models are suitable for the analysis of NMA data, but using random study intercepts requires a strong rationale such as relating treatment effects to study intercepts.

show abstract

Section: Discussionmentioning

confidence: 99%

A comparison of arm‐based and contrast‐based models for network meta‐analysis

White

Turner

Karahalios

et al. 2019

Statistics in Medicine

View full text Add to dashboard Cite

show abstract

“…When clusters vary in size in the population (eg, small versus large general practices), cluster sizes can be seen as realizations of a random variable, and the outcome variable of interest may be related to cluster size (eg, surgeons operating on many patients might have better performances than those operating on fewer patients). If this is the case, then cluster size is said to be informative . Nevalainen et al describe and give practical examples of three data‐generating mechanisms that can lead to informative cluster size.…”

Section: Introductionmentioning

confidence: 99%

“…Briefly, a latent variable (eg, the competence of the surgeon) influences cluster size (eg, the number of patients) and the outcome variable (eg, success of the operation) at the same time; or cluster size affects the outcome variable (eg, surgeons become better by practice); or vice versa, the outcome variable affects cluster size (eg, better surgeons get more referrals). In relation, Seaman et al point out that the standard methods to analyze clustered data, namely, generalized linear mixed models (GLMMs) and generalized estimating equations (GEEs), implicitly assume that cluster size is unrelated to the outcome variable, and discuss different methods to handle informative cluster size for cluster‐specific inference with GLMM and population‐average inference with GEE.…”

Section: Introductionmentioning

confidence: 99%

Relative efficiencies of two‐stage sampling schemes for mean estimation in multilevel populations when cluster size is informative

Innocenti

Candel

Tan

2018

Statistics in Medicine

View full text Add to dashboard Cite

In multilevel populations, there are two types of population means of an outcome variable ie, the average of all individual outcomes ignoring cluster membership and the average of cluster‐specific means. To estimate the first mean, individuals can be sampled directly with simple random sampling or with two‐stage sampling (TSS), that is, sampling clusters first, and then individuals within the sampled clusters. When cluster size varies in the population, three TSS schemes can be considered, ie, sampling clusters with probability proportional to cluster size and then sampling the same number of individuals per cluster; sampling clusters with equal probability and then sampling the same percentage of individuals per cluster; and sampling clusters with equal probability and then sampling the same number of individuals per cluster. Unbiased estimation of the average of all individual outcomes is discussed under each sampling scheme assuming cluster size to be informative. Furthermore, the three TSS schemes are compared in terms of efficiency with each other and with simple random sampling under the constraint of a fixed total sample size. The relative efficiency of the sampling schemes is shown to vary across different cluster size distributions. However, sampling clusters with probability proportional to size is the most efficient TSS scheme for many cluster size distributions. Model‐based and design‐based inference are compared and are shown to give similar results. The results are applied to the distribution of high school size in Italy and the distribution of patient list size for general practices in England.

show abstract

“…The number of totally captured images is still large and this method will lead to redundantly capturing of many leukocytes. Applying clustering algorithm 11,12 on leukocyte nucleus' centroids in leukocyte low power microscopic image based on max-min distance clustering means [13][14][15] can greatly reduce the total number of captured images and the proportion of redundantly capturing. Experimental results indicate that leukocyte image fast scanning method based on maxmin distance clustering has a higher e±ciency.…”

Section: Introductionmentioning

confidence: 99%

A Leukocyte image fast scanning based on max–min distance clustering

Wang

Cao

2016

J. Innov. Opt. Health Sci.

View full text Add to dashboard Cite

A leukocyte image fast scanning method based on max-min distance clustering is proposed. Because of the lower proportion and uneven distribution of leukocytes in human peripheral blood, there will not be any leukocyte in lager quantity of the captured images if we directly scan the blood smear along an ordinary zigzag scanning routine with high power (100 x ) objective. Due to the larger¯eld of view of low power (10 x ) objective, the captured low power blood smear images can be used to locate leukocytes. All of the located positions make up a speci¯c routine, if we scan the blood smear along this routine with high power objective, there will be de¯nitely leukocytes in almost all of the captured images. Considering the number of captured images is still large and some leukocytes may be redundantly captured twice or more, a leukocyte clustering method based on max-min distance clustering is developed to reduce the total number of captured images as well as the number of redundantly captured leukocytes. This method can improve the scanning e±ciency obviously. The experimental results show that the proposed method can shorten scanning time from 8.0-14.0 min to 2.5-4.0 min while extracting 110 nonredundant individual high power leukocyte images.

show abstract

Review of methods for handling confounding by cluster and informative cluster size in clustered data

Cited by 76 publications

References 64 publications

A comparison of arm‐based and contrast‐based models for network meta‐analysis

A comparison of arm‐based and contrast‐based models for network meta‐analysis

Relative efficiencies of two‐stage sampling schemes for mean estimation in multilevel populations when cluster size is informative

A Leukocyte image fast scanning based on max–min distance clustering

Contact Info

Product

Resources

About