Medial Research Network
UK TEL: +44 (0)1908 261 153 | US TEL: +1 (847) 779 7857 | ES TEL: +34 640 36 57 41

Blog Posts

1
Oct
The Future of Patient Identification & Enrolment
Virtual Trials in a Community-Based Model

As discussed in our blog ‘The New Virtual Trial’ we expect trials to become more virtual and adopt the hybrid designs, encompassing virtual visits, Patient Travel Support/Site visits and Home Trial Support visits to patients in their own homes. This will put a lot of trial work into the community and has the potential to drastically reduce site numbers. In our Blog ‘Reducing Site Numbers in Virtual Trials’ we discussed the fact that reducing site numbers is a challenge, but these should not stop us proceeding as it will be a valuable adjunct to what we do today at the least. This blog discusses how we replace one of the key site activities – patient identification – as their numbers reduce. We are moving into the most challenging area of virtual study designs – what I called phase II in my second blog of this series.

Why would we approach this problem today?

Research – what there is of it – indicates that probably only 30% of patients that could enter into studies from a site are ever approached by the site staff about the trial. The average number of patients per site today – according to the PAREXEL source book – is about 7. That means there are probably 23 or 24 available in totality per site. Let’s say our community-based trial tools increase the recruitment of a site from the 7 patients average, to 10. A hypothetical 360 patient study therefore goes from needing 52 sites to needing 36, a reduction of roughly 30%. These extra patients are already there in the site pool, we just need to make sure they are approached by the site. We could probably reasonably safely push the expectation to 14 patients from the site, taking site numbers down to 26. If our aim is further reduction in site numbers we might then target 15 sites as the ideal number to still give adequate geographical coverage, which means each site needs to produce their full 24 patients that are available. Most Project managers would recognise this as a risk and therefore want some from of risk mitigation plan in place.  

Risk Management

The first strand of risk management will include placing more resources in the sites, as even dealing with their original 7 patients tends to strain their capability to resource the visits. Now dealing with 24 is going to be tough.

The second strand of risk management is to generate referrals to the site to ensure they meet their target of 24 patients, by recruiting patients from non-site sources. There are already many ways of doing this in fact, some old and some new. The old guard (tried and tested may be good way of describing them) is to approach brokerage sites on which patients register their interest in trials, or to use prescribing data or something similar to identify where the patients are and direct focused advertising to reach them, or to do more general advertising to patient groups of various sorts. The new guard (bright and shiny new tools – interesting and full of potential but as yet unproven in scale) are social media based using big data analytical tools or focused on recruiting physicians willing to refer patients. The big data systems use data from sources such as Twitter or other social media systems to create big databases against which complex queries can be run to determine if the ‘tweeter’ may be a potential patient for a trial and then to create profiles for these patients that can be used to drive both density mapping to help determine where to place sites and to create focused advertising to reach out to potential patients using tools such as Facebook, Twitter or Google to get customised advertising at low cost. The physician recruitment systems are based on identification of physicians interested in research through various means including big data analysis of the sites and their staff social media output, and then using similar social media or other more traditional methods to reach out to them and engage them in a conversation intended to encourage participation in your trials. Each approach has advantages and disadvantages, essentially based on how many patients they reach with the right condition, how much screening data can you gather to refer ‘good fit’ patients to sites, how interested in trials are the patients you reach and how can you simplify the referral process? 

So far take up of these services seems modest but seems to be growing fast, and many customers and service providers report to us a lack of ability to covert a potential patient into a trial subject.  This is certainly a multifactorial problem, but it does indicate a synergy may be possible with community-based trials. Many of the identified patients must live far away from the sites and so those tools designed to bring distant patients into trial will have a synergistic effect when applied with these other recruitment tools. Essentially, using our service as an example, if we can increase the sites own patient population able to consider participation by 50% we should be able to do the same for the patients identified from other sources.

This itself introduces new challenges to these sites, who now have to look after the patients referred to them as well as coping with the extra work they create by being in the trial – yet more resources as well as systems and communication channels to look after distant patients in volume. The challenges are all practical and logistical, but no less troublesome for that, and there is little innovation in the pipeline designed to address them.

The tools designed to tackle this facet of virtual trials are the most complex and the most challenging as the complexities of the referral process are vast and the ability of patients and service providers to navigate them remains limited. It will be a slow development process, but a certain amount of referral patients is likely to remain a key risk mitigation strategy of trials and it is clear that the community-based tools can support this element as well as the site direct sourced patients, driving development in all areas slowly forward.

Of course – there are other solutions in development all the time. Watch this space.

Posted by:
Toby Heath
Share this post!