Mia Sato Tackney
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Mia Sato Tackney has a total of 145 co-authors. Here are the top co-authors by number of shared publications:
Elizabeth Williamson
Institute of Psychiatry London School of Hygiene and Tropical Medicine London UK
5 shared publications- Elizabeth Williamson
Department of Medical Statistics, London School of Hygiene and Tropical Medicine, London, UK
3 shared publications James Carpenter
Department of Medical Statistics, London School of Hygiene and Tropical Medicine, UK
3 shared publicationsJames R. Carpenter
3 shared publicationsClemence Leyrat
Department of Medical Statistics, London School of Hygiene and Tropical Medicine, London, UK
2 shared publicationsElizabeth Limb
Population Health Research Institute, St George’s, University of London, UK
2 shared publications
Latest Publications
- Open Access
- 9 October 2026
- Published by Elsevier in SSRN Electronic Journal
Accelerating Adoption of Validated Digital Outcome Measures in Clinical Trials: Current Perspectives and Future Directions
Digital measures capture a Meaningful Aspect of Health using sensors in real-world contexts and offer important advantages over traditional in-clinic outcomes that are used in clinical trials. While their use as exploratory outcomes is growing, adoption as primary outcomes remains constrained by operational, methodological and regulatory challenges. To address these, we present a four-stage roadmap for development: from validation to use as exploratory and secondary endpoints, through to first use as a primary endpoint in a confirmatory trial and wider adoption. Key methodological milestones to enable progress through these stages are illustrated with three clinical case studies in Duchenne Muscular Dystrophy, Parkinson’s Disease, and Motor Neurone Disease.
- 9 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessAccelerating Adoption of Validated Digital Outcome Measures in Clinical Trials: Current Perspectives and Future Directions
Digital measures capture a Meaningful Aspect of Health using sensors in real-world contexts and offer important advantages over traditional in-clinic outcomes that are used in clinical trials. While their use as exploratory outcomes is growing, adoption as primary outcomes remains constrained by operational, methodological and regulatory challenges. To address these, we present a four-stage roadmap for development: from validation to use as exploratory and secondary endpoints, through to first use as a primary endpoint in a confirmatory trial and wider adoption. Key methodological milestones to enable progress through these stages are illustrated with three clinical case studies in Duchenne Muscular Dystrophy, Parkinson’s Disease, and Motor Neurone Disease.
- Open Access
- 14 September 2026
- Published by Springer Nature in BMC Medical Research Methodology
Can discussions with patients and the public clarify missing data mechanisms for digital outcome measures?
Background: Analysis of clinical trials with missing data requires statistical assumptions. When novel digital outcome measures are used, it is particularly important to understand trial participants’ experiences with the device. This can illuminate the reasons for and specific patterns of missing data and help inform plausible missing data assumptions. Methods: A patient and public involvement and engagement (PPIE) activity bridged statistical concepts on missing data with lived experiences of individuals who had participated in studies with digital outcome measures. Ten PPIE contributors were involved in the activity, which was structured based on feedback from previous PPIE activities and comprised of three meetings, pre- and post-meeting surveys and shared resources (including an introductory video). The three meetings were aimed to (i) introduce key statistical concepts, discuss expectations and set the scene, (ii) discuss experiences of using digital devices in research studies, and (iii) consolidate learnings and implications. This serves as a structured approach to discussions on missing data which can be used in other settings with digital outcome measures. Results: Reasons for missing data were highly context dependent and varied according to the device, study population and environmental/cultural context. Reasons discussed included operational aspects of the device, unresolved technology issues, and practicalities in daily life such as weather or season affecting comfort and the need to remove devices for religious, exercise or hygiene activities. Previous experience of using digital devices and receiving feedback from devices influenced levels of engagement. Although contributors did not report disengaging with devices in ways directly related to the outcome being measured, informative missingness was considered as important. Conclusions: The PPIE activity was a feasible and valuable approach to exploring patient and public perspectives on the reasons for and patterns of missing data in digital outcome measures. Lived experience may shed light on reasons for missing data that may be overlooked by researchers. Incorporating PPIE discussions within ongoing trials may help inform the plausibility of assumptions, identify candidate auxiliary variables and improve trial conduct. For future PPIE activities, purposeful recruitment and providing different modes of engagement, such as anonymous surveys, are recommended.
- 14 September 2026
- Published by Springer Nature in BMC Medical Research Methodology
Open AccessCan discussions with patients and the public clarify missing data mechanisms for digital outcome measures?
Background: Analysis of clinical trials with missing data requires statistical assumptions. When novel digital outcome measures are used, it is particularly important to understand trial participants’ experiences with the device. This can illuminate the reasons for and specific patterns of missing data and help inform plausible missing data assumptions. Methods: A patient and public involvement and engagement (PPIE) activity bridged statistical concepts on missing data with lived experiences of individuals who had participated in studies with digital outcome measures. Ten PPIE contributors were involved in the activity, which was structured based on feedback from previous PPIE activities and comprised of three meetings, pre- and post-meeting surveys and shared resources (including an introductory video). The three meetings were aimed to (i) introduce key statistical concepts, discuss expectations and set the scene, (ii) discuss experiences of using digital devices in research studies, and (iii) consolidate learnings and implications. This serves as a structured approach to discussions on missing data which can be used in other settings with digital outcome measures. Results: Reasons for missing data were highly context dependent and varied according to the device, study population and environmental/cultural context. Reasons discussed included operational aspects of the device, unresolved technology issues, and practicalities in daily life such as weather or season affecting comfort and the need to remove devices for religious, exercise or hygiene activities. Previous experience of using digital devices and receiving feedback from devices influenced levels of engagement. Although contributors did not report disengaging with devices in ways directly related to the outcome being measured, informative missingness was considered as important. Conclusions: The PPIE activity was a feasible and valuable approach to exploring patient and public perspectives on the reasons for and patterns of missing data in digital outcome measures. Lived experience may shed light on reasons for missing data that may be overlooked by researchers. Incorporating PPIE discussions within ongoing trials may help inform the plausibility of assumptions, identify candidate auxiliary variables and improve trial conduct. For future PPIE activities, purposeful recruitment and providing different modes of engagement, such as anonymous surveys, are recommended.
- Open Access
- 3 August 2026
- Published by Springer Nature in BMC Medical Research Methodology
- Vol. 26 (1)
Comparison of parametric versus machine-learning multiple imputation in clinical trials with missing continuous outcomes
Background: Flexible machine-learning (ML) models to generate imputations within the Multiple Imputation (MI) framework has recently gained traction, particularly in non-randomised observational settings. For randomised controlled trials (RCTs), it is unclear whether ML approaches to MI, in combination with Rubin’s Rules, provide valid inference in terms bias, confidence interval coverage, mean squared error (MSE), Type I error and power of the treatment effect estimator. Methods: We conducted two simulation studies in RCT settings that have an incomplete continuous outcome but fully observed covariates and treatment assignment. We compared Complete Cases, standard MI (MI-norm), MI with predictive mean matching (MI-PMM) and ML-based approaches to MI, including classification and regression trees (MI-CART), Random Forests (MI-RF) and SuperLearner when outcomes are missing completely at random or missing at random conditional on treatment/covariate. The first simulation explored a cross-sectional outcome with non-linear covariate-outcome relationships in the presence/absence of covariate-treatment interactions. The second simulation explored skewed repeated measures, motivated by a trial with digital outcomes. Results: For the cross-sectional simulation without interaction, we found that Complete Cases yielded valid inference; MI-norm performed similarly, except when there is a non-linear covariate-outcome relationship and missingness depends on the covariate. ML approaches led to smaller MSE in specific non-linear settings, but provided unreliable inference for others. MI-PMM, which is the default setting in the package , led to unreliable inference under some settings. In the presence of complex treatment-covariate interactions, performing MI separately by arm, either with MI-norm, MI-RF or MI-CART, provided inference that had comparable or better properties compared to Complete Cases when the analysis model omits the interaction term. In the repeated measures setting, ML approaches to MI and MI-PMM led to bias and under- or over-coverage, particularly when missingness depended on treatment. Conclusion: Based on the simulation findings, Complete Cases and MI-norm are more appropriate than ML approaches to MI for making inference, especially for late phase RCTs where Type I error control is crucial. While ML approaches may provide gains in MSE for complex covariate-outcome relationships, results should be interpreted with the caveat that Rubin’s Rules are not guaranteed to be valid when used with ML imputation methods, and can lead to bias in the estimated effect and/or its standard error.
- 3 August 2026
- Published by Springer Nature in BMC Medical Research Methodology
- Vol. 26 (1)
Open AccessComparison of parametric versus machine-learning multiple imputation in clinical trials with missing continuous outcomes
Background: Flexible machine-learning (ML) models to generate imputations within the Multiple Imputation (MI) framework has recently gained traction, particularly in non-randomised observational settings. For randomised controlled trials (RCTs), it is unclear whether ML approaches to MI, in combination with Rubin’s Rules, provide valid inference in terms bias, confidence interval coverage, mean squared error (MSE), Type I error and power of the treatment effect estimator. Methods: We conducted two simulation studies in RCT settings that have an incomplete continuous outcome but fully observed covariates and treatment assignment. We compared Complete Cases, standard MI (MI-norm), MI with predictive mean matching (MI-PMM) and ML-based approaches to MI, including classification and regression trees (MI-CART), Random Forests (MI-RF) and SuperLearner when outcomes are missing completely at random or missing at random conditional on treatment/covariate. The first simulation explored a cross-sectional outcome with non-linear covariate-outcome relationships in the presence/absence of covariate-treatment interactions. The second simulation explored skewed repeated measures, motivated by a trial with digital outcomes. Results: For the cross-sectional simulation without interaction, we found that Complete Cases yielded valid inference; MI-norm performed similarly, except when there is a non-linear covariate-outcome relationship and missingness depends on the covariate. ML approaches led to smaller MSE in specific non-linear settings, but provided unreliable inference for others. MI-PMM, which is the default setting in the package , led to unreliable inference under some settings. In the presence of complex treatment-covariate interactions, performing MI separately by arm, either with MI-norm, MI-RF or MI-CART, provided inference that had comparable or better properties compared to Complete Cases when the analysis model omits the interaction term. In the repeated measures setting, ML approaches to MI and MI-PMM led to bias and under- or over-coverage, particularly when missingness depended on treatment. Conclusion: Based on the simulation findings, Complete Cases and MI-norm are more appropriate than ML approaches to MI for making inference, especially for late phase RCTs where Type I error control is crucial. While ML approaches may provide gains in MSE for complex covariate-outcome relationships, results should be interpreted with the caveat that Rubin’s Rules are not guaranteed to be valid when used with ML imputation methods, and can lead to bias in the estimated effect and/or its standard error.
- Open Access
DECOVID: A UK Two-Center Harmonized Database of Acute Care Electronic Health Records for COVID-19 Research
The DECOVID database contains harmonized pseudonymized electronic health record (EHR) data on all adult (≥18 years old) patients presenting to two large, digitally mature centers in the United Kingdom between 1 January 2020 and 28 February 2021, with follow-up until at least 28 March 2021. The database was originally developed to support the COVID-19 response but is now available via the PIONEER data hub for researchers to explore a wide range of research questions, including exploratory analyses, risk factor assessment, prediction modeling, and comparative effectiveness studies. Raw data were extracted from local EHRs and transformed into a standardized form (Observational Health Data Sciences and Informatics-Common Data Model version 5.3.1). The database includes 165,420 patients across 256,804 hospital presentations. For these patients, highly granular data are available, including patient demographics, longitudinal vital signs, physiology, treatments, laboratory findings, clinical diagnoses, and outcomes. There are 10,030 patients with COVID-19, of whom 1472 died in hospital.
Open AccessDECOVID: A UK Two-Center Harmonized Database of Acute Care Electronic Health Records for COVID-19 Research
The DECOVID database contains harmonized pseudonymized electronic health record (EHR) data on all adult (≥18 years old) patients presenting to two large, digitally mature centers in the United Kingdom between 1 January 2020 and 28 February 2021, with follow-up until at least 28 March 2021. The database was originally developed to support the COVID-19 response but is now available via the PIONEER data hub for researchers to explore a wide range of research questions, including exploratory analyses, risk factor assessment, prediction modeling, and comparative effectiveness studies. Raw data were extracted from local EHRs and transformed into a standardized form (Observational Health Data Sciences and Informatics-Common Data Model version 5.3.1). The database includes 165,420 patients across 256,804 hospital presentations. For these patients, highly granular data are available, including patient demographics, longitudinal vital signs, physiology, treatments, laboratory findings, clinical diagnoses, and outcomes. There are 10,030 patients with COVID-19, of whom 1472 died in hospital.
- Open Access
Comparison of Parametric versus Machine-learning Multiple Imputation in Clinical Trials with Missing Continuous Outcomes
The use of flexible machine-learning (ML) models to generate imputations of missing data within the framework of Multiple Imputation (MI) has recently gained traction, particularly in observational settings. For randomised controlled trials (RCTs), it is unclear whether ML approaches to MI provide valid inference, and whether they outperform parametric MI approaches under complex data generating mechanisms. We conducted two simulations in RCT settings that have incomplete continuous outcomes but fully observed covariates. We compared Complete Cases, standard MI (MI-norm), MI with predictive mean matching (MI-PMM) and ML-based approaches to MI, including classification and regression trees (MI-CART), Random Forests (MI-RF) and SuperLearner when outcomes are missing completely at random or missing at random conditional on treatment/covariate. The first simulation explored non-linear covariate-outcome relationships in the presence/absence of covariate-treatment interactions. The second simulation explored skewed repeated measures, motivated by a trial with digital outcomes. In the absence of interactions, we found that Complete Cases yields reliable inference; MI-norm performs similarly, except when missingness depends on the covariate. ML approaches can lead to smaller mean squared error than Complete Cases and MI-norm in specific non-linear settings, but provide unreliable inference for others. MI-PMM can lead to unreliable inference in several settings. In the presence of complex treatment-covariate interactions, performing MI separately by arm, either with MI-norm, MI-RF or MI-CART, provides inference that has comparable or with better properties compared to Complete Cases when the analysis model omits the interaction. For ML approaches, we observed unreliable inference in terms of bias in the estimated effect and/or its standard error when Rubin's Rules are implemented.
Open AccessComparison of Parametric versus Machine-learning Multiple Imputation in Clinical Trials with Missing Continuous Outcomes
The use of flexible machine-learning (ML) models to generate imputations of missing data within the framework of Multiple Imputation (MI) has recently gained traction, particularly in observational settings. For randomised controlled trials (RCTs), it is unclear whether ML approaches to MI provide valid inference, and whether they outperform parametric MI approaches under complex data generating mechanisms. We conducted two simulations in RCT settings that have incomplete continuous outcomes but fully observed covariates. We compared Complete Cases, standard MI (MI-norm), MI with predictive mean matching (MI-PMM) and ML-based approaches to MI, including classification and regression trees (MI-CART), Random Forests (MI-RF) and SuperLearner when outcomes are missing completely at random or missing at random conditional on treatment/covariate. The first simulation explored non-linear covariate-outcome relationships in the presence/absence of covariate-treatment interactions. The second simulation explored skewed repeated measures, motivated by a trial with digital outcomes. In the absence of interactions, we found that Complete Cases yields reliable inference; MI-norm performs similarly, except when missingness depends on the covariate. ML approaches can lead to smaller mean squared error than Complete Cases and MI-norm in specific non-linear settings, but provide unreliable inference for others. MI-PMM can lead to unreliable inference in several settings. In the presence of complex treatment-covariate interactions, performing MI separately by arm, either with MI-norm, MI-RF or MI-CART, provides inference that has comparable or with better properties compared to Complete Cases when the analysis model omits the interaction. For ML approaches, we observed unreliable inference in terms of bias in the estimated effect and/or its standard error when Rubin's Rules are implemented.
- Open Access
- 29 August 2025
- Published by SAGE Publications in Statistical Methods in Medical Research
- Vol. 35 (11) , 2367-2388
Implementing response-adaptive designs when responses are missing: Impute or ignore?
Missing data is a widespread issue in clinical trials, but is particularly problematic for digital health interventions where disengagement is common and outcomes are likely to be missing not at random (MNAR). Trials that use response-adaptive designs need to handle missingness online and not simply at the end of the trial. We propose a novel online imputation strategy which allows previous imputations to be re-imputed given updated estimates of success probabilities. We additionally consider: (i) truncation of deterministic algorithms to prevent extreme realised treatment imbalance and (ii) changing the random component of semi-randomised algorithms. Through a simulation study based on a trial for a digital smoking cessation intervention, we illustrate how the strategy for handling missing responses can affect the exploration-exploitation tradeoff and the bias of the estimated success probabilities at the end of the trial. In the settings explored, we found that the exploration-exploitation tradeoff is affected particularly when arms have very different rates of missingness and we identified combinations of response-adaptive designs and missingness strategies that are particularly problematic. Further, we show that estimated success probabilities at the end of the trial can be biased not only due to optimistic sampling, but potentially also due to an MNAR missingness mechanism.
- 29 August 2025
- Published by SAGE Publications in Statistical Methods in Medical Research
- Vol. 35 (11) , 2367-2388
Open AccessImplementing response-adaptive designs when responses are missing: Impute or ignore?
Missing data is a widespread issue in clinical trials, but is particularly problematic for digital health interventions where disengagement is common and outcomes are likely to be missing not at random (MNAR). Trials that use response-adaptive designs need to handle missingness online and not simply at the end of the trial. We propose a novel online imputation strategy which allows previous imputations to be re-imputed given updated estimates of success probabilities. We additionally consider: (i) truncation of deterministic algorithms to prevent extreme realised treatment imbalance and (ii) changing the random component of semi-randomised algorithms. Through a simulation study based on a trial for a digital smoking cessation intervention, we illustrate how the strategy for handling missing responses can affect the exploration-exploitation tradeoff and the bias of the estimated success probabilities at the end of the trial. In the settings explored, we found that the exploration-exploitation tradeoff is affected particularly when arms have very different rates of missingness and we identified combinations of response-adaptive designs and missingness strategies that are particularly problematic. Further, we show that estimated success probabilities at the end of the trial can be biased not only due to optimistic sampling, but potentially also due to an MNAR missingness mechanism.
- Open Access
- 27 September 2024
- Published by Springer Nature in BMC Medicine
- Vol. 22 (1) , 1-20
Unleashing the full potential of digital outcome measures in clinical trials: eight questions that need attention
The use of digital health technologies to measure outcomes in clinical trials opens new opportunities as well as methodological challenges. Digital outcome measures may provide more sensitive and higher-frequency measurements but pose vital statistical challenges around how such outcomes should be defined and validated and how trials incorporating digital outcome measures should be designed and analysed. This article presents eight methodological questions, exploring issues such as the length of measurement period, choice of summary statistic and definition and handling of missing data as well as the potential for new estimands and new analyses to leverage the time series data from digital devices. The impact of key issues highlighted by the eight questions on a primary analysis of a trial are illustrated through a simulation study based on the 2019 Bellerophon INOPulse trial which had time spent in MVPA as a digital outcome measure. These eight questions present broad areas where methodological guidance is needed to enable wider uptake of digital outcome measures in trials.
- 27 September 2024
- Published by Springer Nature in BMC Medicine
- Vol. 22 (1) , 1-20
Open AccessUnleashing the full potential of digital outcome measures in clinical trials: eight questions that need attention
The use of digital health technologies to measure outcomes in clinical trials opens new opportunities as well as methodological challenges. Digital outcome measures may provide more sensitive and higher-frequency measurements but pose vital statistical challenges around how such outcomes should be defined and validated and how trials incorporating digital outcome measures should be designed and analysed. This article presents eight methodological questions, exploring issues such as the length of measurement period, choice of summary statistic and definition and handling of missing data as well as the potential for new estimands and new analyses to leverage the time series data from digital devices. The impact of key issues highlighted by the eight questions on a primary analysis of a trial are illustrated through a simulation study based on the 2019 Bellerophon INOPulse trial which had time spent in MVPA as a digital outcome measure. These eight questions present broad areas where methodological guidance is needed to enable wider uptake of digital outcome measures in trials.
- Open Access
- 3 August 2024
- Published by Springer Nature in Trials
- Vol. 25 (1) , 1-12
Digital endpoints in clinical trials: emerging themes from a multi-stakeholder Knowledge Exchange event
Background: Digital technologies, such as wearable devices and smartphone applications (apps), can enable the decentralisation of clinical trials by measuring endpoints in people’s chosen locations rather than in traditional clinical settings. Digital endpoints can allow high-frequency and sensitive measurements of health outcomes compared to visit-based endpoints which provide an episodic snapshot of a person’s health. However, there are underexplored challenges in this emerging space that require interdisciplinary and cross-sector collaboration. A multi-stakeholder Knowledge Exchange event was organised to facilitate conversations across silos within this research ecosystem. Methods: A survey was sent to an initial list of stakeholders to identify potential discussion topics. Additional stakeholders were identified through iterative discussions on perspectives that needed representation. Co-design meetings with attendees were held to discuss the scope, format and ethos of the event. The event itself featured a cross-disciplinary selection of talks, a panel discussion, small-group discussions facilitated via a rolling seating plan and audience participation via Slido. A transcript was generated from the day, which, together with the output from Slido, provided a record of the day’s discussions. Finally, meetings were held following the event to identify the key challenges for digital endpoints which emerged and reflections and recommendations for dissemination. Results: Several challenges for digital endpoints were identified in the following areas: patient adherence and acceptability; algorithms and software for devices; design, analysis and conduct of clinical trials with digital endpoints; the environmental impact of digital endpoints; and the need for ongoing ethical support. Learnings taken for next generation events include the need to include additional stakeholder perspectives, such as those of funders and regulators, and the need for additional resources and facilitation to allow patient and public contributors to engage meaningfully during the event. Conclusions: The event emphasised the importance of consortium building and highlighted the critical role that collaborative, multi-disciplinary, and cross-sector efforts play in driving innovation in research design and strategic partnership building moving forward. This necessitates enhanced recognition by funders to support multi-stakeholder projects with patient involvement, standardised terminology, and the utilisation of open-source software.
- 3 August 2024
- Published by Springer Nature in Trials
- Vol. 25 (1) , 1-12
Open AccessDigital endpoints in clinical trials: emerging themes from a multi-stakeholder Knowledge Exchange event
Background: Digital technologies, such as wearable devices and smartphone applications (apps), can enable the decentralisation of clinical trials by measuring endpoints in people’s chosen locations rather than in traditional clinical settings. Digital endpoints can allow high-frequency and sensitive measurements of health outcomes compared to visit-based endpoints which provide an episodic snapshot of a person’s health. However, there are underexplored challenges in this emerging space that require interdisciplinary and cross-sector collaboration. A multi-stakeholder Knowledge Exchange event was organised to facilitate conversations across silos within this research ecosystem. Methods: A survey was sent to an initial list of stakeholders to identify potential discussion topics. Additional stakeholders were identified through iterative discussions on perspectives that needed representation. Co-design meetings with attendees were held to discuss the scope, format and ethos of the event. The event itself featured a cross-disciplinary selection of talks, a panel discussion, small-group discussions facilitated via a rolling seating plan and audience participation via Slido. A transcript was generated from the day, which, together with the output from Slido, provided a record of the day’s discussions. Finally, meetings were held following the event to identify the key challenges for digital endpoints which emerged and reflections and recommendations for dissemination. Results: Several challenges for digital endpoints were identified in the following areas: patient adherence and acceptability; algorithms and software for devices; design, analysis and conduct of clinical trials with digital endpoints; the environmental impact of digital endpoints; and the need for ongoing ethical support. Learnings taken for next generation events include the need to include additional stakeholder perspectives, such as those of funders and regulators, and the need for additional resources and facilitation to allow patient and public contributors to engage meaningfully during the event. Conclusions: The event emphasised the importance of consortium building and highlighted the critical role that collaborative, multi-disciplinary, and cross-sector efforts play in driving innovation in research design and strategic partnership building moving forward. This necessitates enhanced recognition by funders to support multi-stakeholder projects with patient involvement, standardised terminology, and the utilisation of open-source software.
- Open Access
- 31 July 2023
- Published by SAGE Publications in Statistical Methods in Medical Research
- Vol. 32 (10) , 1936-1960
Multiple imputation approaches for epoch-level accelerometer data in trials
Clinical trials that investigate physical activity interventions often use accelerometers to measure step count at a very granular level, for example in 5-second epochs. Participants typically wear the accelerometer for a week-long period at baseline, and for one or more week-long follow-up periods after the intervention. The data is aggregated to provide daily or weekly step counts for the primary analysis. Missing data are common as participants may not wear the device as per protocol. Approaches to handling missing data in the literature have defined missingness on the day level using a threshold on daily weartime, which leads to loss of information on the time of day when data are missing. We propose an approach to identifying and classifying missingness at the finer epoch-level and present two approaches to handling missingness using multiple imputation. Firstly, we present a parametric approach which accounts for the number of missing epochs per day. Secondly, we describe a non-parametric approach where missing periods during the day are replaced by donor data from the same person where possible, or data from a different person who is matched on demographic and physical activity-related variables. Our simulation studies show that the non-parametric approach leads to estimates of the effect of treatment that are least biased while maintaining small standard errors. We illustrate the application of these different multiple imputation strategies to the analysis of the 2017 PACE-UP trial. The proposed framework is likely to be applicable to other digital health outcomes and to other wearable devices.
- 31 July 2023
- Published by SAGE Publications in Statistical Methods in Medical Research
- Vol. 32 (10) , 1936-1960
Open AccessMultiple imputation approaches for epoch-level accelerometer data in trials
Clinical trials that investigate physical activity interventions often use accelerometers to measure step count at a very granular level, for example in 5-second epochs. Participants typically wear the accelerometer for a week-long period at baseline, and for one or more week-long follow-up periods after the intervention. The data is aggregated to provide daily or weekly step counts for the primary analysis. Missing data are common as participants may not wear the device as per protocol. Approaches to handling missing data in the literature have defined missingness on the day level using a threshold on daily weartime, which leads to loss of information on the time of day when data are missing. We propose an approach to identifying and classifying missingness at the finer epoch-level and present two approaches to handling missingness using multiple imputation. Firstly, we present a parametric approach which accounts for the number of missing epochs per day. Secondly, we describe a non-parametric approach where missing periods during the day are replaced by donor data from the same person where possible, or data from a different person who is matched on demographic and physical activity-related variables. Our simulation studies show that the non-parametric approach leads to estimates of the effect of treatment that are least biased while maintaining small standard errors. We illustrate the application of these different multiple imputation strategies to the analysis of the 2017 PACE-UP trial. The proposed framework is likely to be applicable to other digital health outcomes and to other wearable devices.
- Open Access
Multiple Imputation Approaches for Epoch-level Accelerometer data in Trials
Clinical trials that investigate interventions on physical activity often use accelerometers to measure step count at a very granular level, often in 5-second epochs. Participants typically wear the accelerometer for a week-long period at baseline, and for one or more week-long follow-up periods after the intervention. The data is usually aggregated to provide daily or weekly step counts for the primary analysis. Missing data are common as participants may not wear the device as per protocol. Approaches to handling missing data in the literature have largely defined missingness on the day level using a threshold on daily wear time, which leads to loss of information on the time of day when data are missing. We propose an approach to identifying and classifying missingness at the finer epoch-level, and then present two approaches to handling missingness. Firstly, we present a parametric approach which takes into account the number of missing epochs per day. Secondly, we describe a non-parametric approach to Multiple Imputation (MI) where missing periods during the day are replaced by donor data from the same person where possible, or data from a different person who is matched on demographic and physical activity-related variables. Our simulation studies comparing these approaches in a number of settings show that the non-parametric approach leads to estimates of the effect of treatment that are least biased while maintaining small standard errors. We illustrate the application of these different MI strategies to the analysis of the 2017 PACE-UP Trial. The proposed framework of classifying missingness and applying MI at the epoch-level is likely to be applicable to a number of different outcomes and data from other wearable devices.
Open AccessMultiple Imputation Approaches for Epoch-level Accelerometer data in Trials
Clinical trials that investigate interventions on physical activity often use accelerometers to measure step count at a very granular level, often in 5-second epochs. Participants typically wear the accelerometer for a week-long period at baseline, and for one or more week-long follow-up periods after the intervention. The data is usually aggregated to provide daily or weekly step counts for the primary analysis. Missing data are common as participants may not wear the device as per protocol. Approaches to handling missing data in the literature have largely defined missingness on the day level using a threshold on daily wear time, which leads to loss of information on the time of day when data are missing. We propose an approach to identifying and classifying missingness at the finer epoch-level, and then present two approaches to handling missingness. Firstly, we present a parametric approach which takes into account the number of missing epochs per day. Secondly, we describe a non-parametric approach to Multiple Imputation (MI) where missing periods during the day are replaced by donor data from the same person where possible, or data from a different person who is matched on demographic and physical activity-related variables. Our simulation studies comparing these approaches in a number of settings show that the non-parametric approach leads to estimates of the effect of treatment that are least biased while maintaining small standard errors. We illustrate the application of these different MI strategies to the analysis of the 2017 PACE-UP Trial. The proposed framework of classifying missingness and applying MI at the epoch-level is likely to be applicable to a number of different outcomes and data from other wearable devices.