Innovative Methods in Trials – Design & Analysis
AusTriM are offering clinical trialists the opportunity to answer important clinical questions using innovative methods through collaboration with one of our Investigators. The successful applicant will receive direct support from the Investigator to either design a new clinical trial or (re)analyse existing trial data.
Applications close 23 September 2022. Please see full details here.
REMAP-CAP and PADDI Trials recognised at 2022 ACTA Trial of the Year Awards
Congratulations to AusTriM Investigators Andrew Forbes and Steve Webb who were recognised for the 2nd year in a row at the ACTA Trial of the Year Awards, celebrating International Clinical Trials Day on May 20th. The awards were presented by Prof Anne Kelso AO, CEO, NHMRC and The Hon Brad Hazzard MP, NSW Minister for Health.
The Randomised, Embedded, Multifactorial Adaptive Platform trial for Community-Acquired Pneumonia trial received the ACTA Trial of the Year Award, with Lisa Higgins accepting the award on behalf of the study team.
Lisa acknowledged Chief Investigator of REMAP-CAP Prof Steve Webb for his ongoing leadership of the trial, which employed a novel design to simultaneously evaluate potential treatments for the care of patients hospitalised in the ICU with COVID-19.
During 2021, the platform reached conclusions regarding a number of interventions including:
- convalescent plasma was unlikely to be of benefit for patients with COVID-19 who require organ support in an intensive care unit
- therapeutic dose of anticoagulation was of benefit for hospitalised patients with COVID-19 who do not require organ support
- immune modulating drugs tocilizumab and sarilumab reduce mortality by 8.5% as well as improving the time to discharge from ICU by about a week; and subsequently, both drugs are equally effective
These results were published and adopted in to clinical guidelines while the platform continues to collect evidence regarding other interventions.
As part of his involvement in the AusTriM network, Steve chairs a network of clinician investigators and biostatisticians working on platform trials across Australia and New Zealand, sharing learnings from existing trials such as REMAP-CAP and providing guidance for investigators on new and establishing trials. He also leads the Adaptive Platform Trials Research stream within the network.
The PADDI Trial
The ACTA-STInG (Statistics in Trials Interest Group) Excellence in Trial Statistics Award is adjudicated by a panel of senior statisticians for the trial that demonstrates exemplary statistical aspects across the life stages of the trial, from trial design and planning, to analysis, reporting and interpretation. This year, the prize was awarded to The Perioperative ADministration of Dexamethasone and Infection (PADDI) trial, with the Trial Statistician, Prof Andrew Forbes, accepting the award on behalf of the PADDI trial group. This marks the second consecutive year that Prof Forbes's work has won this award.
The PADDI trial was a pragmatic randomised non-inferiority trial assessing the safety of 8mg dexamethasone versus placebo in terms of surgical site infection in 8880 patients undergoing non-urgent non-cardiac surgery across 55 centres in 4 countries. The results of the trial were published in the New England Journal of Medicine in 2021.
Andrew Forbes remarked that “Although superficially PADDI was a straightforward stratified individually randomised trial, in practice the non-inferiority design posed many challenges for the design and monitoring of the trial. First, the non-inferiority margin needed to be determined and agreed upon by the large investigator group, because the eventual trial results would ultimately be interpreted with respect to this margin. There were also challenges with respect to determining the asymmetric interim monitoring procedure in which more liberal boundaries were chosen for harm than for non-inferiority. Finally, with the trial likely to be of high clinical interest, the statistical analysis plan needed to be completely comprehensive and replete with sensitivity analyses in order to ensure the methods were transparent and results robust.”
Andrew also acknowledged the clinical leadership of Chief Investigator Prof Tomás Corcoran and the wider Australian and New Zealand College of Anaesthetists (ANZCA) Clinical Trials Network.
Find out about the other award winners from this year’s event on the ACTA website.
AusTriM Clinical Trialist Collaboration Scheme
2021/22 Awardee announced
AusTriM is delighted to be supporting A/Prof Trisha Peel in the design of her research project which will investigate two interventions aimed at reducing postoperative surgical site infections.
In her application, Trisha noted that “surgical site infections (SSIs) are a devastating complication of surgery, leading to significant patient suffering and substantial healthcare costs. It is estimated that over half of all SSIs are preventable. Of concern, infection prevention recommendations are largely informed by low-quality, eminence-based recommendations”.
By using an innovative adaptive design, the study design will simultaneously assess two interventions: i. preoperative bathing with antiseptic cloths versus non-antiseptic cloths and ii. operative surgical site skin preparation with chlorhexidine-alcohol versus iodophor-alcohol.
Adaptive trials have a number of benefits over “standard” randomised clinical trials. They evaluate multiple treatment options; both concurrently, and sequentially by introducing new questions as previous ones are answered. To guide such trials, frequent analyses are performed using Bayesian statistical methods. Extensive pre-trial numerical simulations are undertaken and are updated during the course of the trial as data accumulates, in order to guide the trial’s progress through implementation of stopping rules for benefit, harm, or futility for each treatment.
This requires specialist knowledge of the methodology for planning and implementing such trials and customised development for the specific trial. The complex statistical modelling to support the initial design of this adaptive trial will be overseen by AusTriM Chief Investigator Tom Snelling with work undertaken by Mark Jones, Senior Biostatistician at The University of Sydney.
We congratulate Trisha, second investigator Prof Paul Myles and the wider project team on this new project and look forward to hearing updates throughout 2022.
AusTriM Research Methodology Grant Scheme
2021/22 Recipients announced
Now in its’ second year, AusTriM is again supporting Early-Mid Career Researchers (EMCRs) in developing novel methodologies for the design and analysis of clinical trials. The funded recipients, their coinvestigators, and project titles are:
Dr Robert Mahar, A/Prof Brett Manley, Dr Clare Whitehead and Prof Katherine Lee: Using Bayesian proportional odds to model longitudinal maternal and neonatal outcomes in an adaptive platform trial.
Ms Sabine Braat, Dr Rebecca Harding, Dr Robert Mahar and Prof Julie Simpson: Re-design of intravenous iron program for anaemia in pregnancy in low- and middle-income countries
Dr Anurika De Silva, Ms Sabine Braat, Prof Katherine Lee and Prof Julie Simpson: Novel methods for missing data in trials affected by COVID-19
We look forward to hearing updates on these projects throughout 2022 at our monthly Work in Progress meetings, and Trial Analysis and Platform Trials special interest group seminars.
Three PhD Conferrals – Congratulations Drs McLeod,
Jachno and Grantham!
We are delighted to announce PhD conferrals for three of our students.
Kim M Jachno, PhD
Analysis Of Treatment Effects And Event Rates That Change Over Time In Clinical Trials
Main Supervisor: Prof Rory Wolfe
Kim will continue to be involved in the AusTriM network, based at Monash University School of Public Health and Preventive Medicine.
Charlie McLeod, PhD
Improving the Methodology for PAtient-Centred Trials in Cystic Fibrosis (IMPACT CF)
Main Supervisor: Prof Tom Snelling
Charlie will continue to be involved with the AusTriM network, and shortly commence her postdoctoral role at Telethon Kids Institute, WA.
Kelsey Grantham, PhD
Statistical Methods For The Design And Analysis Of Longitudinal Cluster Randomized Trials
Main Supervisor: Prof. Andrew Forbes
Kelsey will continue her research in the area of cluster randomised trials and shortly commence her postdoctoral role at Monash University SPHPM, supervised by AusTriM Associate Investigator Jess Kasza.
Innovative Methods in Trials – Design & Analysis
AusTriM is offering clinical trialists the opportunity to receive advice and support to answer important clinical questions using innovative methods. The successful applicant / team can expect us to collaborate with them to either i. design a new clinical trial (Stream A) or ii. analyse existing trial data (Stream B).
What's on offer:
STREAM A: TRIAL DESIGN
Do you have a clinical question and wish to design a trial using innovative methodologies to answer it?
STREAM B: DATA ANALYSIS
Do you have existing trial data to be (re)analysed using innovative approaches?
Our Investigator team can collaborate with you to utilise methodologies such as the following:
- Adaptive platform trials
- Multi-arm multi-stage (MAMS) designs
- Cluster randomised crossover trials
- Stepped wedge trials (e.g. with time-to-event endpoints)
- Partially clustered trials
- Compliance-adjusted analysis in randomised trials
- Methods for missing data
- Composite endpoints
- Treatment mechanisms and surrogate outcomes
Please click here (PDF) for further details and how to apply.
PEPTIC Trial wins ACTA-STInG Excellence in Trial Statistics award, REMAP-CAP Corticosteroid Domain commended
The Australian Clinical Trials Alliance (ACTA) marked International Clinical Trials Day with their National Tribute and Award Ceremony. AusTriM Chief Investigators Prof Andrew Forbes and Prof Steve Webb were recognised for their involvement in the PEPTIC and REMAP-CAP trials, respectively.
The Proton pump inhibitor vs. histamine-2 rEceptor blockers for ulcer Prophylaxis Therapy in the Intensive Care unit (PEPTIC) Trial received the ACTA-STInG (Statistics in Trials Interest Group) Excellence in Trial Statistics award, with the Trial Statistician, Prof Andrew Forbes, accepting the award on behalf of the PEPTIC trial group. The trial was also recognised as a finalist in the Trial of the Year category.
The trial used an innovative cluster crossover methodology to compare two strategies for stress ulcer prophylaxis in critically ill patients admitted to ICU. Utilising registry and other routinely collected data, the trial randomised 26,982 patients from a total of 50 ICUs across Australia, Canada, New Zealand, Ireland and the United Kingdom. It was the largest clinical trial to ever be conducted in the intensive care setting, yet at a budget of only $500,000 USD, or 1/10th of a typical large critical care trial; largely because of the efficiency of the design. The trial results were published in JAMA in January 2020.
Prof Forbes developed theoretical results and simulation work for the cluster crossover design of this trial. The unique methodology has the potential to address clinical questions that were previously considered logistically or financially infeasible, and has provided the impetus for current and future research in this area of innovative trial design.
Andrew said: “In PEPTIC, it turned out that the careful stratification of sites into regional and temporal batches resulted in the trial having almost the same precision as a 27,000 patient individually randomised trial. This in itself is quite remarkable, and indicates the value of careful planning, together of course with a healthy dose of good luck.”
Through the AusTriM network, Prof Forbes and collaborators are now developing a suite of the most efficient ‘incomplete factorial’ cluster randomised crossover designs and their sample size requirements, together with associated tutorials and software tools developed in R and Stata. AusTriM aims to disseminate these findings through close collaboration with clinical trialists, statisticians, clinical trial networks and ACTA. For more detail on our current projects please refer to our Research and Publications pages.
The Randomised, Embedded, Multifactorial Adaptive Platform trial for Community-Acquired Pneumonia (REMAP-CAP) trial – Corticosteroid Domain received commendations in both the Excellence in Trial Statistics and Trial of the Year Award categories.
REMAP-CAP is an ongoing Australian-led ongoing international adaptive platform trial which aims to improve outcomes for critically ill patients admitted to the ICU with pneumonia. With the coronavirus pandemic, this platform has been adapted to find the best treatment strategies for COVID-19 patients receiving intensive care in hospitals worldwide. The corticosteroid domain of the platform was one of the first to be incorporated early in 2020.
The scale and design of the trial allow it to assess the effectiveness of a range of treatments already routinely used in clinical practice, as well as the flexibility to quickly incorporate new treatments as they demonstrate promise and “drop” treatments that either fail to perform, or when there is a loss of equipoise, as happened with the corticosteroid domain.
When data released from the RECOVERY Trial demonstrated a benefit of corticosteroid therapy, recruitment of patients to the REMAP-CAP corticosteroid domain was halted the very next day. REMAP-CAP subsequently demonstrated a 93% and 80% probability of benefit on the primary outcome of organ support-free days at day 21 for the fixed-duration and the shock-dependent course, respectively, when compared with no assigned corticosteroids. The results of this were published in JAMA in September 2020.
AusTriM Methodology Research Seed Funding Grant Scheme
As part of our mission to support early/mid-career researchers (E/MCRs) within the network, we have launched a scheme to provide seed funding for research projects aimed at either: (a) developing new design and analysis methods for clinical trials; or (b) implementation of innovative methods into clinical trials.
Applications were judged by a panel of senior researchers according to criteria based on those used by the NHMRC (significance, innovativeness, alignment, feasibility and track record/career).
Congratulations to the awardees – we look forward to progress updates on your work throughout 2021:
- Dr Thao Le, Prof Rory Wolfe, Prof Stephane Heritier, A/Prof Ronald Geskus – Handling missing disease information due to death in trial endpoints that need two visits to diagnose
- Dr Julie Marsh, Prof Kate Lee, Dr Kaushala Jayawardana, Mr Michael Dymock – BRACE Trial redesign project
- Dr Rhys Bowden, Prof Andrew Forbes, Dr Jessica Kasza – Guiding the choice of within-cluster correlation structure for cluster randomised crossover and stepped wedge trials: a comprehensive simulation study
- Dr Robert Mahar, Prof Julie Simpson – A SMART design with Bayesian adaptive randomisation: optimising multi-stage prophylaxis and treatment regimens for immunocompromised patients
- Dr Lisa Yelland, Dr Thomas Sullivan, Prof Andrew Forbes, Prof Katherine Lee – Comparison of Analysis Methods for Addressing Imperfect Stratification
Steve Webb recognised with nomination for The Australian newspaper's Australian of the Year Award
Prof Steve Webb, AusTriM lead investigator, was nominated for his leadership of the Randomised, Embedded, Multi-factorial, Adaptive Platform Trial for Community-Acquired Pneumonia (REMAP-CAP).
REMAP-CAP is an Australian-led ongoing international adaptive platform trial which aims to improve outcomes for critically ill patients admitted to the ICU with pneumonia. With the coronavirus pandemic, this platform has been adapted to find the best treatment strategies for COVID-19 patients receiving intensive care in hospitals worldwide.
Prof Webb leads the global consortium of nearly 100 international investigators which is active at almost 300 hospitals and is rapidly approaching 10,000 randomisations in 5,000 patients with suspected or proven COVID infections. The international team, under Prof Webb's chairmanship, has discovered that using two existing arthritis treatments can save one in 12 seriously ill patients (noting that this finding is still under peer-review). The platform has also reported the effectiveness of corticosteroids as well as made announcements relating to the treatment effect of antiviral agents, anticoagulation, and convalescent plasma. The platform continues to evaluate the effect of antiplatelet agents, vitamin C, statin therapy, and additional immune modulators.
This is a massive accolade and important recognition of his coordinated efforts and leadership to uncover some of the best intensive care treatments for COVID-19, potentially saving hundreds of thousands of lives.
ARC Discovery Project
Congratulations to Dr Jessica Kasza and Professor Andrew Forbes who were awarded an ARC Discovery grant (2021-25) totalling $420,000 for their project "Increasing the efficiency and interpretability of stepped wedge cluster randomised trials".
Their collaborators for this project are Professor Richard Hooper, Queen Mary University of London, Professor James Hughes (University of Washington), and Dr Schadrac Agbla (University of Liverpool).
Andrew Forbes elected as a Fellow of AAHMS
Professor Andrew Forbes has been elected as a Fellow of the Australian Academy of Health and Medical Sciences (AAHMS).
In welcoming the newly elected Fellows at the 2020 Scientific Meeting on 14 October, the Academy’s President, Professor Ingrid Scheffer, acknowledged that "Fellowship is granted in recognition of a significant and sustained contribution to health and medical research, demonstrating leadership on both the Australian and international stage".
Andrew was recognised for his leadership in biostatistics, reflected in his development of innovative statistical methods to solve practical issues in health and medical research studies, as well as collaboration on major practice-changing research in substantive clinical and public health areas.
Andrew is recognised as an international expert in the design of cluster crossover trials. More information on his research can be found on his profile.
AusTriM featured in Research Australia's INSPIRE magazine
Innovative trials including REMAP-CAP, ASCOT and CLARITY are being used to examine the efficacy of potential treatments for COVID-19.
Read the full article here.
NHMRC Investigator grant
Congratulations to Associate Investigator Professor Julie Simpson who was awarded a NHMRC Investigator grant (2021-25) totalling $2,163,220 for her research "Optimising treatment and prevention strategies to accelerate malaria elimination".
AusTriM working to develop critical COVID-19 trials
Researchers within our network are playing a crucial part in national and international trials to protect the healthcare workforce; evaluate treatment options for patients and build real-time clinical decision support in to health systems. Read more about these trials below.
BCG vaccination to Reduce the impact of COVID-19 in Australian healthcare workers following Coronavirus Exposure (BRACE) Trial
Professor Kate Lee, Trial Statistician
Healthcare workers are at increased risk of contracting COVID-19, which has significant implications for the health care system during a pandemic. The BCG vaccine, generally used to protect against tuberculosis, also boosts immunity to other infections. The purpose of the BRACE trial is to determine if BCG vaccination reduces the incidence and severity of COVID-19 in Australian healthcare workers during the 2020 pandemic. This Phase 3, two-group, multicentre, open-label, randomized, controlled trial aims to recruit 4,170 health care workers in Australian hospitals. Primary results are expected in October 2020, and full study results are expected in March 2022.
For more information see the trial website: mcri.edu.au/BRACE
Randomized, Embedded, Multifactorial Adaptive Platform trial for Community-Acquired Pneumonia (REMAP-CAP) adds COVID-19 domains
Professor Steve Webb, Intensive Care Physician and Principal Investigator
REMAP-CAP is an ongoing international adaptive platform trial which aims to improve outcomes for critically ill patients admitted to the ICU with pneumonia. The novel design of the REMAP-CAP trial enables it to adapt in the event of a pandemic to evaluate the most relevant treatment options. In early 2020, three additional domains were added to the trial protocol to treat COVID-19 patients, with over 300 patients randomised to date.
You can read more about REMAP-CAP and listen to an interview with Steve here: remapcap.org/media
ASCOT (Australasian COVID-19 Trial) rolls out across hospitals in Australia and New Zealand
Professor Tom Snelling, Infectious Disease Physician and Investigator
ASCOT will utilise response adaptive randomisation to allocate COVID-19 patients being treated in hospital to a range of supportive therapies. Current treatments being investigated are standard care (no intervention), kaletra and hydroxychloroquine. By utilising Bayesian statistical methods, additional treatment arms may be added as new treatments become available, just as existing arms may be dropped if they are shown to not be effective.
Read more about why the trial will continue to investigate hydroxychloroquine despite recent controversy in the media.
NHMRC Ideas Grant
Congratulations to Chief Investigator Professor Andrew Forbes who was awarded a 2019 NHMRC Ideas Grant totalling $480,350 for his project ‘Pragmatic cluster randomised trials: Bridging the gap between theory and practice’.
Best Paper and Publication Prize
Congratulations to AusTriM Associate Investigator Dr Jessica Kasza who has been awarded the 2019 Alfred Research Alliance EMCR Public Health and Clinical Research Best Paper Award and the 2019 John McNeil Early Career Researcher Publication Prize for Public Health Research for her paper jointly authored with Professor Andrew Forbes:
Kasza J, Forbes AB. Inference for the treatment effect in multiple-period cluster randomised trials when random effect correlation structure is misspecified. Statistical Methods in Medical Research. 2019: 28(10–11), 3112–3122. doi: 10.1177/0962280218797151
AusTriM Australian Trials Methodology Conference 2021
The full program is now available for our 2021 Conference which is being held fully online over two days in December. The meeting program showcases cutting-edge trial design and analysis approaches in a way that is thought-provoking yet still accessible to biostatisticians, methodologists and trialists (across any clinical discipline) alike.
We are thrilled to announce our lineup of invited speakers Marcel Wolbers (Roche, Switzerland), Stephen Senn (Consultant Statistician, UK), Susan Murphy (Harvard University, USA), Elaine Pascoe (University of Queensland), Tom Snelling (University of Sydney), Dimitris Rizopoulos (Erasmus Medical Center, The Netherlands), Suzie Cro (Imperial College London, UK), Ian White (University College London, UK), Phillip Westgate (University of Kentucky, USA), Charles Weijer (Western University, Canada) and Fan Li (Yale University, USA).
The conference program also includes four contributed sessions from 20 local and international speakers on the topics of adaptive trial design, issues in the analysis of trials and innovations in clinical trial design.
There will be ample opportunity for attendees to engage with speakers and fellow delegates in our networking lounge and during our Social event on the Monday evening. While our in-person meeting will have to wait, we look forward to networking with and learning from our colleagues and invited speakers in this digital space.
Registration closes on Friday 3 December – Standard: $200 / Student: $100
WORKSHOP: MODERN CONCEPTS IN CLINICAL TRIALS
Led by investigators Katherine Lee and Julie Marsh, this series of 'hands-on' workshops covered adaptive designs and the estimand framework.
22–25 February 2021, half-day online workshops
DATA AND SAFETY MONITORING BOARDS (DSMB) – STATISTICAL PRINCIPLES
Professor Ian Marschner
24 November 2020, webinar
DATA AND SAFETY MONITORING BOARDS (DSMB) – HOW TO PLAN WELL AND BE PREPARED FOR THE UNEXPECTED
Professor Carmel Hawley
10 November 2020, webinar
ADAPTIVE PLATFORM TRIALS – AVOIDING REGRET IN TRIAL DESIGN
Professor Steve Webb
11 November 2020, webinar
TRIAL DESIGNS FOR REGISTRY-BASED TRIALS
Professor Andrew Forbes
19 May 2020, webinar
BAYESIAN ADAPTIVE RANDOMISED CLINICAL TRIALS WORKSHOP
Dr Lorenzo Trippa and Dr Steffen Ventz
27–28 February 2020, Melbourne / 5–6 March 2020, Sydney
CURRENT DEVELOPMENTS IN CLUSTER-RANDOMISED TRIALS AND STEPPED WEDGE DESIGNS
Professor Andrew Forbes, Professor Karla Hemming, Associate Professor Monica Taljaard
21–22 November 2019, Melbourne
Call for PhD applicants
Students interested in undertaking a PhD in trials methodology are encouraged to contact firstname.lastname@example.org to register their interest. Current opportunities are listed below, or see our research page for further information on our areas of study.
Students within the AusTriM network will benefit from ongoing training and mentorship, paid professional development opportunities, peer networking across Australia and may be eligible to receive a top-up scholarship of up to $6,000 per annum.
Students may be based at one of our activity nodes at Monash University, Murdoch Children’s Research Institute, Telethon Kids Institute and Adaptive Health Intelligence, University of Western Australia, South Australian Health and Medical Research Institute, the NHMRC Clinical Trials Centre (CTC) at the University of Sydney, University of Queensland and the University of Melbourne.
Expressions of interest should be forwarded to email@example.com and include a copy of your:
ii. Academic Transcript
Dealing with missing outcome data in randomised trials with extended follow up
Supervisors: Professor Katherine Lee (Murdoch Children’s Research Institute, Melbourne) and Dr Thomas Sullivan (South Australian Health and Medical Research Institute, Adelaide)
Location: Melbourne or Adelaide
Background: Missing outcome data can pose a serious problem in randomised trials, requiring careful consideration during analysis to avoid biasing conclusions. The challenges of missing outcome data are especially prominent in “extended follow-up” studies, where an additional follow-up period is commenced after the protocol defined completion of a trial to learn about longer-term impacts of the intervention. Extended follow-up studies often involve additional eligibility restrictions and consent processes, which can further limit the amount of data available for analysis. While the method of multiple imputation has been widely recommended for addressing missing outcome data in randomised trials, its application in extended follow-up studies is not well understood.
- To review the literature on methods of multiple imputation for randomised trials and assess their suitability and limitations for trials with extended follow-up.
- To explore how the inclusion of randomised participants ruled ineligible or not consenting to extended follow-up in an imputation model affects the validity of treatment effect estimates.
- To evaluate the ability of multiple imputation to address large blocks of missing outcome data, as occurs when participants fail to provide any data during extended follow-up, compared with alternative approaches such as complete case analysis or using multiple imputation in combination with inverse probability weighting.
- To develop approaches and provide guidance on conducting sensitivity analyses in extended follow-up studies when the probability of missing data is suspected to depend on unobserved data.
Optimising dynamic treatment regimens using sequential multiple assignment randomised trial data with missing data
Supervisors: Dr Robert Mahar (Murdoch Children’s Research Institute/University of Melbourne), Prof Katherine Lee (Murdoch Children’s Research Institute), Prof Julie Simpson (University of Melbourne)
Background: Clinicians often face multi-stage and dynamic decisions when treating patients with either chronic or progressive medical conditions. Dynamic treatment regimens formalise such sequential decision problems. Sequential multiple assignment randomised trials (SMARTs) are clinical studies that randomise patients to different treatments over time, which provides data that can be used to optimise dynamic treatment regimens. But repeatedly randomising patients to treatments can lead to complicated patterns of missing data, a problem with few existing SMART-specific solutions. This PhD project will develop and evaluate statistical methods to optimise dynamic treatment regimens using data from hypothetical SMART designs with problematic missing data.
- To review the literature on methods to impute missing data for randomised trials and assess their suitability and limitations for SMARTs.
- To formalise patterns of SMART missing data using directed acyclic graphs.
- To evaluate the ability of different imputation methods to deal with different patterns of missingness compared with alternative approaches such as complete case analysis.
- To develop approaches and provide guidance on conducting sensitivity analyses in SMARTs when the probability of missing data is suspected to depend on unobserved data.
Handling missing data in repeatedly measured outcomes when assessing change over time
Supervisors: Prof Katherine Lee (Murdoch Children’s Research Institute / The University of Melbourne), Dr Rheanna Mainzer (Murdoch Children’s Research Institute)
Location: Clinical Epidemiology and Biostatistics Unit, Murdoch Children’s Research Institute
Background: Missing data are common in medical research, particularly in studies with multiple waves of follow-up. One particular challenge is how to handle missing data in analyses that involve the change in a variable over time where data are missing at one or more time points for some individuals. Analysis methods specifically designed for longitudinal data, such as mixed effects models and generalised estimation equations, allow all individuals with the outcome measured at any time point to be included in the analysis, but would exclude participants who have missing data at all time points. Multiple imputation (MI) is a popular method for handling missing data, including in longitudinal studies, which would enable all participants to be included in the analysis. This PhD project will develop and evaluate statistical methods for analysing change over time in the presence of missing data.
- To review the literature on methods for analysing change over time in the presence of missing data.
- To evaluate different methods for handling missing data in studies where the change over time is of interest.
We welcome EOIs for the following project:
Accelerated failure time models in health studies
Supervisors: Professor Rory Wolfe (Monash University, School of Public Health and Preventive Medicine)