King's College London

Research portal

Measurement error, time lag, unmeasured confounding: Considerations for longitudinal estimation of the effect of the mediation ‘b path’ in randomised clinical trials

Research output: Contribution to journalArticle

Standard

Measurement error, time lag, unmeasured confounding : Considerations for longitudinal estimation of the effect of the mediation ‘b path’ in randomised clinical trials. / Goldsmith, KA; Chalder, T; White, PD; Sharpe, M; Pickles, A.

In: Statistical Methods in Medical Research, Vol. 27, No. 6, 01.06.2018, p. 1615-1633.

Research output: Contribution to journalArticle

Harvard

Goldsmith, KA, Chalder, T, White, PD, Sharpe, M & Pickles, A 2018, 'Measurement error, time lag, unmeasured confounding: Considerations for longitudinal estimation of the effect of the mediation ‘b path’ in randomised clinical trials', Statistical Methods in Medical Research, vol. 27, no. 6, pp. 1615-1633. https://doi.org/10.1177/0962280216666111

APA

Goldsmith, KA., Chalder, T., White, PD., Sharpe, M., & Pickles, A. (2018). Measurement error, time lag, unmeasured confounding: Considerations for longitudinal estimation of the effect of the mediation ‘b path’ in randomised clinical trials. Statistical Methods in Medical Research, 27(6), 1615-1633. https://doi.org/10.1177/0962280216666111

Vancouver

Goldsmith KA, Chalder T, White PD, Sharpe M, Pickles A. Measurement error, time lag, unmeasured confounding: Considerations for longitudinal estimation of the effect of the mediation ‘b path’ in randomised clinical trials. Statistical Methods in Medical Research. 2018 Jun 1;27(6):1615-1633. https://doi.org/10.1177/0962280216666111

Author

Goldsmith, KA ; Chalder, T ; White, PD ; Sharpe, M ; Pickles, A. / Measurement error, time lag, unmeasured confounding : Considerations for longitudinal estimation of the effect of the mediation ‘b path’ in randomised clinical trials. In: Statistical Methods in Medical Research. 2018 ; Vol. 27, No. 6. pp. 1615-1633.

Bibtex Download

@article{5133e2ddbfc84ef981dd3cbb63e6f1bf,
title = "Measurement error, time lag, unmeasured confounding: Considerations for longitudinal estimation of the effect of the mediation ‘b path’ in randomised clinical trials",
abstract = "Clinical trials are expensive and time-consuming and so should also be used to study how treatments work. This would allow evaluation of theoretical treatment models and refinement and improvement of treatments. Treatment processes can be studied using mediation analysis. Randomised treatment makes some of the assumptions of mediation models plausible, but the mediator – outcome relationship remains one that can be subject to bias. In addition, mediation is assumed to be a temporally ordered longitudinal process, but most mediation studies to date have been cross-sectional and unable to explore this assumption. This study used longitudinal structural equation modelling of mediator and outcome measurements from the PACE trial of rehabilitative treatments for chronic fatigue syndrome (ISRCTN 54285094) to address these issues. In particular, autoregressive and simplex models were used to study measurement error in the mediator, different time lags in the mediator – outcome relationship, unmeasured confounding of the mediator and outcome, and the assumption of a constant mediator – outcome relationship over time. Results showed that allowing for measurement error and unmeasured confounding were important. Concurrent rather than lagged mediator – outcome effects were more consistent with the data, possibly due to the wide spacing of measurements. Assuming a constant mediatoroutcome relationship over time increased precision.",
author = "KA Goldsmith and T Chalder and PD White and M Sharpe and A Pickles",
year = "2018",
month = "6",
day = "1",
doi = "10.1177/0962280216666111",
language = "English",
volume = "27",
pages = "1615--1633",
journal = "Statistical Methods in Medical Research",
issn = "0962-2802",
publisher = "SAGE Publications Ltd",
number = "6",

}

RIS (suitable for import to EndNote) Download

TY - JOUR

T1 - Measurement error, time lag, unmeasured confounding

T2 - Considerations for longitudinal estimation of the effect of the mediation ‘b path’ in randomised clinical trials

AU - Goldsmith, KA

AU - Chalder, T

AU - White, PD

AU - Sharpe, M

AU - Pickles, A

PY - 2018/6/1

Y1 - 2018/6/1

N2 - Clinical trials are expensive and time-consuming and so should also be used to study how treatments work. This would allow evaluation of theoretical treatment models and refinement and improvement of treatments. Treatment processes can be studied using mediation analysis. Randomised treatment makes some of the assumptions of mediation models plausible, but the mediator – outcome relationship remains one that can be subject to bias. In addition, mediation is assumed to be a temporally ordered longitudinal process, but most mediation studies to date have been cross-sectional and unable to explore this assumption. This study used longitudinal structural equation modelling of mediator and outcome measurements from the PACE trial of rehabilitative treatments for chronic fatigue syndrome (ISRCTN 54285094) to address these issues. In particular, autoregressive and simplex models were used to study measurement error in the mediator, different time lags in the mediator – outcome relationship, unmeasured confounding of the mediator and outcome, and the assumption of a constant mediator – outcome relationship over time. Results showed that allowing for measurement error and unmeasured confounding were important. Concurrent rather than lagged mediator – outcome effects were more consistent with the data, possibly due to the wide spacing of measurements. Assuming a constant mediatoroutcome relationship over time increased precision.

AB - Clinical trials are expensive and time-consuming and so should also be used to study how treatments work. This would allow evaluation of theoretical treatment models and refinement and improvement of treatments. Treatment processes can be studied using mediation analysis. Randomised treatment makes some of the assumptions of mediation models plausible, but the mediator – outcome relationship remains one that can be subject to bias. In addition, mediation is assumed to be a temporally ordered longitudinal process, but most mediation studies to date have been cross-sectional and unable to explore this assumption. This study used longitudinal structural equation modelling of mediator and outcome measurements from the PACE trial of rehabilitative treatments for chronic fatigue syndrome (ISRCTN 54285094) to address these issues. In particular, autoregressive and simplex models were used to study measurement error in the mediator, different time lags in the mediator – outcome relationship, unmeasured confounding of the mediator and outcome, and the assumption of a constant mediator – outcome relationship over time. Results showed that allowing for measurement error and unmeasured confounding were important. Concurrent rather than lagged mediator – outcome effects were more consistent with the data, possibly due to the wide spacing of measurements. Assuming a constant mediatoroutcome relationship over time increased precision.

UR - http://www.scopus.com/inward/record.url?scp=85047171208&partnerID=8YFLogxK

U2 - 10.1177/0962280216666111

DO - 10.1177/0962280216666111

M3 - Article

VL - 27

SP - 1615

EP - 1633

JO - Statistical Methods in Medical Research

JF - Statistical Methods in Medical Research

SN - 0962-2802

IS - 6

ER -

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454