↓ Skip to main content

A comparison of methods to adjust for continuous covariates in the analysis of randomised trials

Overview of attention for article published in BMC Medical Research Methodology, April 2016
Altmetric Badge

About this Attention Score

  • In the top 25% of all research outputs scored by Altmetric
  • High Attention Score compared to outputs of the same age (87th percentile)

Mentioned by

twitter
25 tweeters
facebook
1 Facebook page

Citations

dimensions_citation
14 Dimensions

Readers on

mendeley
61 Mendeley
citeulike
2 CiteULike
Title
A comparison of methods to adjust for continuous covariates in the analysis of randomised trials
Published in
BMC Medical Research Methodology, April 2016
DOI 10.1186/s12874-016-0141-3
Pubmed ID
Authors

Brennan C. Kahan, Helen Rushton, Tim P. Morris, Rhian M. Daniel

Abstract

Although covariate adjustment in the analysis of randomised trials can be beneficial, adjustment for continuous covariates is complicated by the fact that the association between covariate and outcome must be specified. Misspecification of this association can lead to reduced power, and potentially incorrect conclusions regarding treatment efficacy. We compared several methods of adjustment to determine which is best when the association between covariate and outcome is unknown. We assessed (a) dichotomisation or categorisation; (b) assuming a linear association with outcome; (c) using fractional polynomials with one (FP1) or two (FP2) polynomial terms; and (d) using restricted cubic splines with 3 or 5 knots. We evaluated each method using simulation and through a re-analysis of trial datasets. Methods which kept covariates as continuous typically had higher power than methods which used categorisation. Dichotomisation, categorisation, and assuming a linear association all led to large reductions in power when the true association was non-linear. FP2 models and restricted cubic splines with 3 or 5 knots performed best overall. For the analysis of randomised trials we recommend (1) adjusting for continuous covariates even if their association with outcome is unknown; (2) keeping covariates as continuous; and (3) using fractional polynomials with two polynomial terms or restricted cubic splines with 3 to 5 knots when a linear association is in doubt.

Twitter Demographics

The data shown below were collected from the profiles of 25 tweeters who shared this research output. Click here to find out more about how the information was compiled.

Mendeley readers

The data shown below were compiled from readership statistics for 61 Mendeley readers of this research output. Click here to see the associated Mendeley record.

Geographical breakdown

Country Count As %
Switzerland 2 3%
Spain 1 2%
Sweden 1 2%
United Kingdom 1 2%
Unknown 56 92%

Demographic breakdown

Readers by professional status Count As %
Researcher 17 28%
Student > Ph. D. Student 12 20%
Professor > Associate Professor 6 10%
Student > Master 4 7%
Professor 4 7%
Other 14 23%
Unknown 4 7%
Readers by discipline Count As %
Medicine and Dentistry 18 30%
Mathematics 11 18%
Biochemistry, Genetics and Molecular Biology 4 7%
Agricultural and Biological Sciences 3 5%
Social Sciences 2 3%
Other 9 15%
Unknown 14 23%

Attention Score in Context

This research output has an Altmetric Attention Score of 13. This is our high-level measure of the quality and quantity of online attention that it has received. This Attention Score, as well as the ranking and number of research outputs shown below, was calculated when the research output was last mentioned on 09 July 2019.
All research outputs
#1,272,436
of 14,093,618 outputs
Outputs from BMC Medical Research Methodology
#202
of 1,293 outputs
Outputs of similar age
#33,126
of 263,649 outputs
Outputs of similar age from BMC Medical Research Methodology
#1
of 1 outputs
Altmetric has tracked 14,093,618 research outputs across all sources so far. Compared to these this one has done particularly well and is in the 90th percentile: it's in the top 10% of all research outputs ever tracked by Altmetric.
So far Altmetric has tracked 1,293 research outputs from this source. They typically receive more attention than average, with a mean Attention Score of 9.2. This one has done well, scoring higher than 84% of its peers.
Older research outputs will score higher simply because they've had more time to accumulate mentions. To account for age we can compare this Altmetric Attention Score to the 263,649 tracked outputs that were published within six weeks on either side of this one in any source. This one has done well, scoring higher than 87% of its contemporaries.
We're also able to compare this research output to 1 others from the same source and published within six weeks on either side of this one. This one has scored higher than all of them