Statistical reanalysis of vascular event outcomes in primary and secondary vascular prevention trials

Lisa J Woodhouse; Alan A Montgomery; Jonathan Mant; Barry R Davis; Ale Algra; Jean-Louis Mas; Jan A Staessen; Lutgarde Thijs; Andrew Tonkin; Adrienne Kirby; Stuart J Pocock; John Chalmers; Graeme J Hankey; J David Spence; Peter Sandercock; Hans-Christoph Diener; Shinichiro Uchiyama; Nikola Sprigg; Philip M Bath

doi:10.1186/s12874-021-01388-6

Statistical reanalysis of vascular event outcomes in primary and secondary vascular prevention trials

Lisa J Woodhouse, Alan A Montgomery, Jonathan Mant, Barry R Davis, Ale Algra, Jean-Louis Mas, Jan A Staessen, Lutgarde Thijs, Andrew Tonkin, Adrienne Kirby, Stuart J Pocock, John Chalmers, Graeme J Hankey, J David Spence, Peter Sandercock, Hans-Christoph Diener, Shinichiro Uchiyama, Nikola Sprigg, Philip M Bath

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

Abstract

BACKGROUND: Vascular prevention trials typically use dichotomous event outcomes although this may be inefficient statistically and gives no indication of event severity. We assessed whether ordinal outcomes would be more efficient and how to best analyse them.

METHODS: Chief investigators of vascular prevention randomised controlled trials that showed evidence of either benefit or harm, or were included in a systematic review that overall showed benefit or harm, shared individual participant data from their trials. Ordered categorical versions of vascular event outcomes (such as stroke and myocardial infarction) were analysed using 15 statistical techniques and their results then ranked, with the result with the smallest p-value given the smallest rank. Friedman and Duncan's multiple range tests were performed to assess differences between tests by comparing the average ranks for each statistical test.

RESULTS: Data from 35 trials (254,223 participants) were shared with the collaboration. 13 trials had more than two treatment arms, resulting in 59 comparisons. Analysis approaches (Mann Whitney U, ordinal logistic regression, multiple regression, bootstrapping) that used ordinal outcome data had a smaller average rank and therefore appeared to be more efficient statistically than those that analysed the original binary outcomes.

CONCLUSIONS: Ordinal vascular outcome measures appear to be more efficient statistically than binary outcomes and provide information on the severity of event. We suggest a potential role for using ordinal outcomes in vascular prevention trials.

Original language	English
Article number	218
Journal	BMC Medical Research Methodology
Volume	21
Issue number	1
DOIs	https://doi.org/10.1186/s12874-021-01388-6
Publication status	Published - Dec 2021

Access to Document

10.1186/s12874-021-01388-6Licence: CC BY

Cite this

Woodhouse, L. J., Montgomery, A. A., Mant, J., Davis, B. R., Algra, A., Mas, J-L., Staessen, J. A., Thijs, L., Tonkin, A., Kirby, A., Pocock, S. J., Chalmers, J., Hankey, G. J., Spence, J. D., Sandercock, P., Diener, H-C., Uchiyama, S., Sprigg, N., & Bath, P. M. (2021). Statistical reanalysis of vascular event outcomes in primary and secondary vascular prevention trials. BMC Medical Research Methodology, 21(1), [218]. https://doi.org/10.1186/s12874-021-01388-6

@article{10c1d69915654d7ab58c2341ca85e2cd,

title = "Statistical reanalysis of vascular event outcomes in primary and secondary vascular prevention trials",

abstract = "BACKGROUND: Vascular prevention trials typically use dichotomous event outcomes although this may be inefficient statistically and gives no indication of event severity. We assessed whether ordinal outcomes would be more efficient and how to best analyse them.METHODS: Chief investigators of vascular prevention randomised controlled trials that showed evidence of either benefit or harm, or were included in a systematic review that overall showed benefit or harm, shared individual participant data from their trials. Ordered categorical versions of vascular event outcomes (such as stroke and myocardial infarction) were analysed using 15 statistical techniques and their results then ranked, with the result with the smallest p-value given the smallest rank. Friedman and Duncan's multiple range tests were performed to assess differences between tests by comparing the average ranks for each statistical test.RESULTS: Data from 35 trials (254,223 participants) were shared with the collaboration. 13 trials had more than two treatment arms, resulting in 59 comparisons. Analysis approaches (Mann Whitney U, ordinal logistic regression, multiple regression, bootstrapping) that used ordinal outcome data had a smaller average rank and therefore appeared to be more efficient statistically than those that analysed the original binary outcomes.CONCLUSIONS: Ordinal vascular outcome measures appear to be more efficient statistically than binary outcomes and provide information on the severity of event. We suggest a potential role for using ordinal outcomes in vascular prevention trials.",

author = "Woodhouse, {Lisa J} and Montgomery, {Alan A} and Jonathan Mant and Davis, {Barry R} and Ale Algra and Jean-Louis Mas and Staessen, {Jan A} and Lutgarde Thijs and Andrew Tonkin and Adrienne Kirby and Pocock, {Stuart J} and John Chalmers and Hankey, {Graeme J} and Spence, {J David} and Peter Sandercock and Hans-Christoph Diener and Shinichiro Uchiyama and Nikola Sprigg and Bath, {Philip M}",

year = "2021",

month = dec,

doi = "10.1186/s12874-021-01388-6",

language = "English",

volume = "21",

journal = "BMC Medical Research Methodology",

issn = "1471-2288",

publisher = "BioMed Central",

number = "1",

}

Woodhouse, LJ, Montgomery, AA, Mant, J, Davis, BR, Algra, A, Mas, J-L, Staessen, JA, Thijs, L, Tonkin, A, Kirby, A, Pocock, SJ, Chalmers, J, Hankey, GJ, Spence, JD, Sandercock, P, Diener, H-C, Uchiyama, S, Sprigg, N & Bath, PM 2021, 'Statistical reanalysis of vascular event outcomes in primary and secondary vascular prevention trials', BMC Medical Research Methodology, vol. 21, no. 1, 218. https://doi.org/10.1186/s12874-021-01388-6

TY - JOUR

T1 - Statistical reanalysis of vascular event outcomes in primary and secondary vascular prevention trials

AU - Woodhouse, Lisa J

AU - Montgomery, Alan A

AU - Mant, Jonathan

AU - Davis, Barry R

AU - Algra, Ale

AU - Mas, Jean-Louis

AU - Staessen, Jan A

AU - Thijs, Lutgarde

AU - Tonkin, Andrew

AU - Kirby, Adrienne

AU - Pocock, Stuart J

AU - Chalmers, John

AU - Hankey, Graeme J

AU - Spence, J David

AU - Sandercock, Peter

AU - Diener, Hans-Christoph

AU - Uchiyama, Shinichiro

AU - Sprigg, Nikola

AU - Bath, Philip M

PY - 2021/12

Y1 - 2021/12

N2 - BACKGROUND: Vascular prevention trials typically use dichotomous event outcomes although this may be inefficient statistically and gives no indication of event severity. We assessed whether ordinal outcomes would be more efficient and how to best analyse them.METHODS: Chief investigators of vascular prevention randomised controlled trials that showed evidence of either benefit or harm, or were included in a systematic review that overall showed benefit or harm, shared individual participant data from their trials. Ordered categorical versions of vascular event outcomes (such as stroke and myocardial infarction) were analysed using 15 statistical techniques and their results then ranked, with the result with the smallest p-value given the smallest rank. Friedman and Duncan's multiple range tests were performed to assess differences between tests by comparing the average ranks for each statistical test.RESULTS: Data from 35 trials (254,223 participants) were shared with the collaboration. 13 trials had more than two treatment arms, resulting in 59 comparisons. Analysis approaches (Mann Whitney U, ordinal logistic regression, multiple regression, bootstrapping) that used ordinal outcome data had a smaller average rank and therefore appeared to be more efficient statistically than those that analysed the original binary outcomes.CONCLUSIONS: Ordinal vascular outcome measures appear to be more efficient statistically than binary outcomes and provide information on the severity of event. We suggest a potential role for using ordinal outcomes in vascular prevention trials.

AB - BACKGROUND: Vascular prevention trials typically use dichotomous event outcomes although this may be inefficient statistically and gives no indication of event severity. We assessed whether ordinal outcomes would be more efficient and how to best analyse them.METHODS: Chief investigators of vascular prevention randomised controlled trials that showed evidence of either benefit or harm, or were included in a systematic review that overall showed benefit or harm, shared individual participant data from their trials. Ordered categorical versions of vascular event outcomes (such as stroke and myocardial infarction) were analysed using 15 statistical techniques and their results then ranked, with the result with the smallest p-value given the smallest rank. Friedman and Duncan's multiple range tests were performed to assess differences between tests by comparing the average ranks for each statistical test.RESULTS: Data from 35 trials (254,223 participants) were shared with the collaboration. 13 trials had more than two treatment arms, resulting in 59 comparisons. Analysis approaches (Mann Whitney U, ordinal logistic regression, multiple regression, bootstrapping) that used ordinal outcome data had a smaller average rank and therefore appeared to be more efficient statistically than those that analysed the original binary outcomes.CONCLUSIONS: Ordinal vascular outcome measures appear to be more efficient statistically than binary outcomes and provide information on the severity of event. We suggest a potential role for using ordinal outcomes in vascular prevention trials.

U2 - 10.1186/s12874-021-01388-6

DO - 10.1186/s12874-021-01388-6

M3 - Article

C2 - 34657596

VL - 21

JO - BMC Medical Research Methodology

JF - BMC Medical Research Methodology

SN - 1471-2288

IS - 1

M1 - 218

ER -

Statistical reanalysis of vascular event outcomes in primary and secondary vascular prevention trials

Abstract

Access to Document

Fingerprint

Cite this