Teaching Programmes

MBChB Year 2 Integrated Problems (statistics)

Master in Public Health (ESRM and Advanced Statistics)

Potential PhD opportunities

I am always keen to hear from potential PhD students, both those with and without funding. From time to time funded PhD studentships are available. Some suggested PhD programes of research are listed below. However, I would also be happy to hear from perspective students  to work with me and other collaborators to develop a new funding proposal, either with access to their own funds, or to submit to a funding body. If any of these ideas interest you, do get in touch. We are working in a new research field, and there are lots of exciting opportunities. My work is in both statistics and general methodology of trials and evidence based medicine, so you don’t have to be a statistician to apply.

Examples of projects:

What is the best way to analyse the data from a SW-CRT? What are the implications of the generalised estimating equation approach and the random effects approach on bias and coverage?  How do these modelling methods compare with within cluster comparisons and within step comparisons?

What is the risk of selection bias or baseline imbalance in a SW-CRT and how can this be mitigated? Does the stepped wedge design lessen the potential for baseline imbalance compared to the parallel design? Does pairing or stratification improve baseline imbalance?

Development of Stata add on functions for the design and analysis of stepped wedge studies

Past projects

What is the current quality of reporting of the SW-CRT? Do recently reported SW-CRTs confirm to the consort extension for parallel cluster trials? How are recent SW-CRTs powered and analysed (with specific emphasis on time effects and appropriate design effect); and do they report the design features such as the number and timing of steps?

Research and reporting methods for the stepped wedge cluster randomised controlled trial

The stepped‐wedge cluster randomised trial (SW‐CRT) is a novel research study design that  is increasingly being used in the evaluation of service delivery type interventions. Recent systematic reviews have demonstrated that use of this study design is clearly on the increase. In the early history of cluster randomised trials it was not unusual to see trial results which were published without allowing for the clustering, and so results which were overly precise. Because the SW‐CRT is a new study design, it is possible that the analysis and reporting of these trials are not adequately reported. Reporting guidelines specific to SWCRTs do not exist, but here we recommend some minor additions or modifications that can be made to the cluster consort guidelines until specific guidelines for SW‐CRTs are formalised. 

One day work shop: Stepped wedge cluster RCT Roadshow 
For information on upcoming dates please contact Karla: k.hemming@bham.ac.uk  

Collaborators: Karla works with an international group of people all interested and actively researching into the stepped wedge design.
This includes: Monica Taljaard, Associate Professor, Epidemiology and Community Medicine University of Ottawa http://www.ohri.ca/profile/mtaljaard and Terry Haines, Monash University http://www.med.monash.edu.au/physio/staff/haines.html

Selected Publications

Hemming K, Haines TP, Chilton PJ, Girling AJ, Lilford RJ. The stepped wedge cluster randomised trial: rationale, design, analysis, and reporting. BMJ (Clinical research ed). 2015;350:h391.

Hemming K, Lilford R, Girling AJ. Stepped-wedge cluster randomised controlled trials: a generic framework including parallel and multiple-level designs. Statistics in medicine. 2015;34(2):181-96.

K Hemming AJ Girling T Haines R Lilford Protocol: Consort extension to stepped wedge cluster ran-domised controlled trials Equator network  http://www.equator-network.org/wp-content/uploads/2009/02/Consort-SW-Protocol-V1.pdf

Hemming K, Girling A. A menu-driven facility for power and detectable-difference calculations in stepped-wedge cluster-randomized trials. Stata Journal. 2014;14(2):363-80.

Hemming K, Girling A. The efficiency of stepped wedge vs. cluster randomized trials: stepped wedge studies do not always require a smaller sample size. Journal of clinical epidemiology. 2013;66(12):1427-8.

Hemming K, Girling A, Martin J, Bond SJ. Stepped wedge cluster randomized trials are efficient and provide a method of evaluation without which some interventions would not be evaluated. Journal of clinical epidemiology. 2013;66(9):1058-9.

Hemming K, Girling AJ, Sitch AJ, Marsh J, Lilford RJ. Sample size calculations for cluster randomised controlled trials with a fixed number of clusters. BMC medical research methodology. 2011;11:102.

Hemming K, Marsh J. A menu-driven facility for sample-size calculations in cluster randomized controlled trials. Stata Journal. 2013;13(1):114-35.