Countries being used for our WebCollect studies
Argentina
Australia
Austria
Belgium
Brazil
Canada
Chile
Czech Republic
Estonia
Finland
France
Germany
Hungary
Ireland
Israel
Italy
Lithuania
Mexico
Netherlands
Portugal
Slovakia
South Africa
Spain
Switzerland
United Kingdom
United States
October 2001
Web-based Studies:
Set-up for Support and Success
Remote Data Capture has been around for quite a few years now - in various forms and permutations. Some companies have made a success of using remote electronic systems, and these have usually been the ones who approach the whole issue with a very open mind regarding processes. If new systems were squashed into old processes, then the result was often a more time-consuming, frustrating study, with no real gains made in terms of efficiency.
In recent years, there have been big advances made in Internet technology and security, which have led in turn to rapid progress in development of web-based data capture tools. Having the tools is one part of the equation, but persuading people to use them is something else! Most of us are fairly resistant to change - even if we like to think we are not - and when the pressure is on we tend to fall back on tried and trusted methods as it makes us feel more secure. We know what we are doing, and can make a reasonably good, educated prediction about the likely outcome. Thus, with new systems it is absolutely vital to promote good confidence by imparting as much knowledge as possible, and providing good back up.
We are going to look at the critical factors for setting up a successful web-based study:
Planning and Team-building
Training
User/Site Assessment
Support Systems
Contingency Plans
Planning & Team-building
An integrated, co-operating team works together for best results - if it is the first time you have worked together, you really do need a truly joint approach. The word "partnership" may be overused, but you will get best results in an open culture.
It should be noted that extra care needs to be taken when designing/building eCRFs as changing them later in a study is more difficult than changing a standard database. Therefore, having all the team together deciding the most protocol accurate, user-friendly (clinical), buildable (database administrators), thoroughly clean-able (data management), easily analys-able (statistics) eCRF is a great benefit. Time spent here is well worth it in the long run, but be careful not to go round in circles on things that are personal preferences - stick to those pertinent to the study.
Involving all services in the set-up is a good way to allay concerns, and to pick up good ideas. For example, the supplies people may be able to get a report from the web-based system which lets them know when stocks are getting low at site, and re-supply can be triggered.
We do need to remember that this is still a clinical trial, even though we are using new technology - it is easy to lose focus. The Project Plan may be in a different format, but it still contains all the usual items. In addition, it will include the eCRF plan and associated programs, and plans for support services, assessments, vendor liaison, and anything else useful.
It is worth double-checking that your vendor agreement matches your project plan, and have a method in place to communicate changes between the two.
The portal that everyone uses for accessing the database can be customised to hold your study updates, newsletters, etc. Access can be controlled so that certain areas are only available to certain groups within the study.
Training
Training is an integral, compulsory part of our own WebCollect service. This is partly to reflect the requirements of 21 CFR Part 11 - which needs written proof of training - and partly to ensure that everyone using the chosen system is confident in its use. It is a good idea to have a degree of one-on-one training if possible, as some people find it difficult to speak up and ask questions when in groups.
We are lucky enough to have a dedicated training group who handle both the internal training of staff at Quintiles, such as the CRAs, and also training at Investigator Meetings and on site, if necessary. It is very beneficial, as there are some fundamental changes in roles when using the web-based tools.
We have worked closely with our vendors to develop training modules, which have been well received. Training all staff together at an Investigator's meeting can also be good for relationship building and for gaining buy-in as everyone hears the same story.
User/Site Assessment
Users and sites must be thoroughly assessed for hardware, software, expertise, training needs, time, ISP.
Having said that, it is imperative that the technical competence of a site is not the governing factor in their participation - that should remain their clinical expertise and suitability, as always. Ideally, the system should have an off-line option, which allows sites to participate in the study, even if Internet connections are poor.
Most sites are happy to join web-based studies as more and more are realising that it will save time and effort. There is also the attraction of being involved in something new and innovative, especially in Europe and the less developed countries.
However, don't underestimate the amount of investment needed! Anywhere up to 50% of sites will need something in an average study - from full kit to an ISP or ISDN line.
We have a standard questionnaire, which we use to assess all new sites, which was developed with the help of our vendors, and is regularly updated.
This step is absolutely critical - it cannot be emphasised enough.
The table left gives you some idea of the types of countries being used for our WebCollect studies - more are being added all the time. As our industry looks for treatment naïve patients and bigger patient populations, we are moving into less developed countries - most of whom have some good connectivity and hardware, but not necessarily in the hospitals and doctor's surgeries.
Support Systems
As an adjunct to good training, all staff involved must have somewhere to turn if they run into difficulties. Ideally, one freephone number should connect them with all the help they need for any aspect of a study. They can use this as much or as little as necessary, but it is always there, 24 hours every day.
Agreements with system providers, Internet Service providers, our own IT, hardware sources and medical backup allow us to direct calls to our Helpdesk appropriately. We do recognise that we cannot be expert in all aspects, so we have brought in those who are expert to ensure all areas are covered.
It is worthwhile checking with your Helpdesk provider that you have the full coverage that you need, with access to all the languages involved in your study.
Our Helpdesk also plays a pivotal role in the User/Site Assessments, as they are available to go through the questionnaires with the sites in their own language (or using real-time translation services), thus improving the quality and completeness of the information we receive.
Look out for non-study related issues: we had one investigator calling us because he ran into problems with the games he had installed on his study laptop...!!
Contingency Plans
Contingency plans for potential disaster are always included in our Project Plans, but there are some items in a web-based trial, which would not occur in a paper one.
Back-up tapes may be taken more regularly, and if sites are using an off-line version of a system, they need to be taught to either transfer data promptly and/or take back-ups of their own.
There may be some concerns over the business stability of relatively new vendors, so measures can be put in place to minimise risks for all. For example, if a company does go down, Quintiles has rights to a copy of the software to ensure that current trials can be completed
Some pharma companies like to have back-up paper copies of the CRF printed in small volumes - if you combine paper and EDC, you do tend to lose the time advantages, as you are limited by your slowest factors, but it may be useful for a short-term emergency.
And to finish, a few highlights and a few words of caution from our own experience:
Highlights:
We have had successful audits in four countries with no major findings for web-based studies.
Most sites do complete their data within a few days of the visit, although some need encouragement!
Recruitment can be effectively tracked using just the web.
Data validation query process well accepted by everyone, once they have grasped what they are doing.
Be careful!
Speed: people have strange expectations about Internet speed; try to encourage them to think it will be like using the Internet normally, not like working on their hard drive.
IT: there is enormous variation between sites, even within a country; very thorough assessment is necessary.
Hardware: don't under estimate requirements.
Training: make sure all relevant staff attend and have plan in place for new staff (including site staff).
Changing eCRF: can be complex and expensive.
Helpdesk: protocol amendments, eCRF changes, staff changes can lead to a big increase in calls.
Evolutionary process: reduced "comfort zone" for the sites - may need extra motivation from the project team.
In summary:
Planning:
ensures the team know what needs to be done and who is responsible.
Training:
increases compliance and acceptance.
Assessments:
appropriate tools available.
Support:
help of whatever type whenever needed.
Contingency:
hope we never need it!
Selina Sibbald,
Quintiles, Scotland Ltd