June 29, 2021
Design & Statistics
Pilot and feasibility studies for pragmatic cluster randomised trials
May 2021 – In Grand Rounds 17, Claire Chan presents on the growing number of studies described as pilot or feasibility studies, and describes the challenges in defining and designing these studies.
Speaker
Claire Chan, MSc
Statistician
Institute of Population Health Sciences (IPHS)
Queen Mary University of London
Webinar Recording
Learning Objectives
- To understand what pilot and feasibility studies are and how to design them
- To understand some of the special considerations for pilot and feasibility studies in advance of a pragmatic trial
- To understand some of the special considerations for cluster randomised trials
Podcast
s
Podcast Transcript: Download the transcript here.
Additional Resources
Hey and Taljaard co-authors on review of pragmatic clinical trials
March 28, 2021
IMPACT Collaboratory members Spencer Hey, PhD and Monica Taljaard, PhD are among the authors of a review of pragmatic clinical trials recently published online in the Journal of Clinical Epidemiology and covered in a blog post from the US GRADE network. The first author of the study is Stuart G. Nicholls, PhD, of the Ottawa Hospital Research Institute.
The researchers identified 4,337 eligible trials in their review of data collected from MEDLINE, Web of Science, ClinicalTrials.gov and full text between January 2014 and April 2019. They found diversity in research areas, but only 22% of the trials identified were labeled as pragmatic in the title, abstract or full text of the document. In addition, the authors estimated one-fifth of trials under-accrued by at least 15%. They concluded that there is a need to improve reporting of pragmatic trials and quality of trial registry data. Under accrual remains a challenge in pragmatic RCTs despite calls for more streamlined recruitment approaches.
The US GRADE Network blog developed an infographic to share the findings of the review.
Goldfeld examines constrained randomization to evaulate the vaccine rollout in nursing homes
Keith Goldfeld, member of the Executive Committee of the IMPACT Design & Statistics Core (DSC), is part of the IMPACT Collaboratory study examining the rollout of COVID-19 vaccine in nursing homes. He recently authored a blog post describing the statistical approach to the study.
The effort involves a cluster randomized trial (CRT) to evaluate the effectiveness of a wide-ranging outreach program designed to encourage nursing home residents to get their shots. The focus is on facilities that have relatively high proportions of African-American and Latinx residents, because these facilities have been among the least successful in the past in convincing residents to get vaccinated. The outcome measure of the trial, which will be measured at the individual level, will be the probability of vaccination within 5 weeks of being available at the nursing home.
Read the full blog post at: https://www.rdatagen.net/post/2020-12-22-constrained-randomization-to-evaulate-the-vaccine-rollout-in-nursing-homes/
Bynum and Travison presentations from NIA ADRD Summit now available
Julie Bynum, MD, MPH, Core Leader of the IMPACT Technical Data Core and Thomas Travison, PhD, Associate Core Leader of the Design and Statistics Core, presented at the National Institute on Aging Summit Virtual Meeting Series: 2020 National Research Summit on Care, Services, and Supports for Persons with Dementia and Their Caregiver this summer. The slides and presentation videos are now available at: https://bit.ly/3nreloY.
Goldfeld blog examines a hurdle model for COVID-19 infections in nursing homes
A hurdle model is a modified count model that also assumes a relatively high frequency of zeros, but is set up as a two-stage data generation process rather than as a mixture distribution. In his most recent blog post, Keith Goldfeld, DrPH, MS, MPA of the IMPACT Design and Statistics Core discusses using a hurdle model in the context of planning a new study to address COVID-19 infection rates in nursing homes using much more aggressive testing strategies. Read the full blog post here.
Goldfeld Blog Post: Consider a permutation test for a small pilot study
Check out the latest blog post from Keith Goldfeld, DrPH, MS, MPA, executive committee member of the IMPACT Design and Statistics Core. Goldfeld considers the use of a permutation test for a small pilot study.
Link: https://www.rdatagen.net/post/permutation-test-for-a-covid-19-pilot-nursing-home-study/
New blog post on clustered randomized trials and the design effect
The latest blog post from Keith Goldfeld, DrPH, MS, MPA, member of the executive committee of the IMPACT Collaboratory Design and Statistics Core, discusses how the design effect could provide additional insight into cluster randomized trials in his most recent blog post. The ideas in this blog post were motivated in part by Goldfeld’s work with the IMPACT Collaboratory.
Goldfeld discusses analyzing data from a stepped-wedge designed cluster randomized trial in a real-world context
Keith Goldfeld, DrPH, MS, MPA, member of the executive committee of the IMPACT Collaboratory Design and Statistics Core, recently discussed his work with the IMPACT Collaboratory on his blog, ouR data generation.
Goldfeld wrote about how to analyze data from a stepped-wedge designed cluster randomized trial. He said that this design can be an important tool to evaluate interventions in a real-world context, including research for the IMPACT Collaboratory in dementia and Alzheimers disease.
Read the full blog entry here: https://www.rdatagen.net/post/analyzing-the-open-cohort-stepped-wedge-trial-with-binary-outcomes/