SAS Art Competition 2016 - The Results
Once again, we have been very impressed by the variety of submissions sent in by statisticians and SAS programmers. There can only be one winner however...
Some considerations when planning the statistical approach to an IPD meta-analysis of time-to-event data
We recently undertook a meta-analysis of time-to-adverse-event data across 12 randomized controlled trials in chronic obstructive pulmonary disease and share a few lessons learned when planning and implementing this type of analyses. Meta-analysis of individual patient data (IPD) offers a number of advantages over the traditional meta-analysis of aggregate data - in particular the ability to reliably estimate covariate and interaction effects. IPD can be modelled in a â€˜one-stage" analysis (fitting a single model to all trials simultaneously), or the trials can be modelled separately and summary statistics analysed using standard meta-analytic techniques, known as a â€˜two-stage" approach. One-stage models are appealing when IPD are available, but two-stage approaches should not be discounted as they are simpler to implement (particularly if assuming random effects) and any loss of power appears to be small (Stewart et al. 2012, Fisher et al. 2011).
In our analysis we aimed to establish three things:
Specialist providers praised at PCT
PHASTAR recently attended the annual Partnerships in Clinical Trials conference in Vienna. This conference brings together the clinical trial community including pharma, biotech and medical device professionals, CROs, site investigators and patients, providing a platform for open discussion, debate and knowledge exchange. The key themes for this year's conference were patient centricity, disruptive innovation and collaboration.
One of the recurring discussion points throughout the conference was the merits of full-service versus functional outsourcing. Pharma companies have been outsourcing for over 25 years but there still remains a level of dissatisfaction among sponsors. Many of the roundtable discussions debated the reasons behind this dissatisfaction and highlighted both good and bad experiences. A common emerging theme was the relative success of functional outsourcing where a specialist provider is more open to adopting innovative approaches or have expertise/capacity not available in the sponsor organisation. This is in addition to the growing need for a flexible resourcing strategy which adapts to changes in company portfolios and priorities.
PHASTAR push for CDASH standards
Most sponsor companies have recognised the need for, and have long been working to, data collection and processing standards. Great. But when you think that these standards are typically specific to a company, a therapeutic area or the project team conducting the study, then the amount of "standards" in play is quite considerable. Now that the FDA is requesting the submission of clinical trial data in common CDISC format, in order to ease the burden of their review process, the various formats employed and held on to so closely and for so long are no longer viable.
The current trend is to convert resulting datasets into the required SDTM format, either during the conduct of an ongoing study, or even once the study has ended for older legacy studies. As a CRO, we receive data from multiple clients, in multiple therapeutic areas and all can attest that the level of programming effort required for this task varies according to the format in which the data are obtained. In instances where data needs to be pooled and the data from each of the studies has been provided in different formats, this effort needs to be duplicated as the programs will need to be tailored to each of the formats. This is obviously very time consuming.