Time to switch?

In light of the increasing discussions on patient-centricity within clinical trials, patient-reported outcomes (PROs) are being used more widely in research. They can offer huge insight into the patient perspective of living with the disease, including the symptom burden and impact on their quality of life. Patients themselves have also been calling for more high-quality patient-outcome data to be made available. Similarly, a number of national and international organisations and policymakers, such as the American Society of Clinical Oncology (ASCO) in the US, the European Medical Association (EMA) in Europe, and the National Institute for Health and Care Excellence (NICE) in the UK have also demonstrated support for this idea.

Derek Kyte, lecturer in health research methods at the University of Birmingham, has conducted a lot of research into the use of PROs for clinical trials and routine practice. These two different areas can feed into each other.

“In oncology, for example, capturing PRO trial data, alongside survival and progression data, allows future patients and clinicians to gain a sense of how they might feel on a given therapy, as well as how long they might live,” says Kyte. “This information is incredibly important for helping patients make more informed decisions about their management, both at the point of diagnosis and beyond.”

Kevin Crawford, associate director of Tenax Therapeutics, is also enthusiastic about the use of PROs in clinical trials. This is because they provide unique information that can supplement the other data being collected over the course of a study. “The benefits include really being able to dive a little bit deeper into primarily secondary outcomes for clinical trials, secondary end points,” says Crawford. “A lot of times, although studies will not use this data for registration reasons, they’ll use them as additional talking points. It’s really an unbiased look from a patient’s point of view.”

Electronic patient-reported outcomes (ePROs), where data is collected via multiple platforms including PC, tablet, smartphone or telephone voice recognition, can offer further benefits for clinical trials. For this reason, their usage is rapidly increasing. Evidence suggests their use dramatically increases compliance, reducing avoidable missing data.

For instance, a recent multicentre trial conducted in the US found over 90% compliance with the use of ePROs, which was supplemented with central coordinator backup. This high level of compliance was sustained, with over 70% at the six-month post-treatment follow-up. These results were particularly promising because the associated time commitment from site research associates and central coordinators was modest. Results of this trial were reported in the Journal of Clinical Oncology.

Using ePROs can also offer unique benefits to patients. “I think it gives a sense of empowerment to the patients, especially if you hand them a device that they’re able to take home, that they can enter this information at their leisure,” says Crawford. “It gives them the ability to be really a part of the trial, not just a ‘subject’ in the trial.”

Collecting this data electronically is also highly valuable for sponsors. “We can get data back quicker, as long as we’re taking a look at it, and if we have a migration into our electronic case-report form,” explains Crawford. “When you have paper, you have the issue of transcription errors occurring, so electronic is much better for that case. I also think it’s scalable.”

Any reason not too?

There are particular trials where, instead of using ePROs, more challenging and more traditional methods may be preferable. “We did a clinical trial on people with digital ulcers, where there is a lack of vascularity to the end of the fingertips,” says Crawford. “If you’re punching on a screen that may actually be a bad thing and patients would rather hold a pen or a pencil. Or if you were doing a migraine trial and patients didn’t want to look at a screen, paper might be much easier for that patient.”

Ensuring the measure is user-friendly is another important consideration. “If you want to provide multiple-choice answers, and the last option is off the screen and the patients have to scroll up or down to see number six, you’re going to miss that question 80% of the time,” says Crawford. “I can see a lot of times where if you didn’t standardise the tool that they were doing it on and it was just an app that you allowed them to download, you could have a very different set of data.”

From China to the US – the value of standardisation

Standardisation is also hugely important when using ePRO in global trials. “Infrastructure-wise, it’s challenging when trying to get the ePRO into different markets,” explains Crawford. “If you’re doing a trial and you’re based in the US, and you want to access the internet in China, for example. Standardising that, making sure that the code is the same, that it’s read the same and the instructions are the same, is imperative.”

The format must also be agile. “Having a data capture system in place that is flexible enough to cater to the target trial demographic is important,” says Kyte. “This should include various options for formatting or mode of administration – font versions, appropriate colour schemes, or voice recognition for the visually impaired, for example.”

Testing ePRO measures on a small scale can be hugely helpful in ensuring that they produce useful data. “I would definitely recommend piloting it in some way to better understand, do some user testing to see if it works,” says Crawford.

Despite the huge potential of ePRO for providing unique and insightful data for clinical trials, there remains resistance in the industry.

“I was at a conference, just last month, and it was 50/50 on the interest in ePRO,” says Crawford. “I think a lot of people are still more comfortable with the idea of handing out paper.”

Best practice makes perfect

When implementing ePRO, there are a number of best practices to ensure that they generate valuable data as part of a trial.

“The first thing that comes to mind is to go with a recommendation,” says Crawford. “I would speak to peers in the industry, find somebody that has used them before.”

When using a vendor, asking a lot of questions is important to establish whether they are a good fit for a particular trial. “I would vet the company a bit,” says Crawford. “How long have they been doing ePRO? Is it on call 24/7? Who do we call if the ePRO goes down? How quickly do you replace equipment? How often do you update your databases? Do things go offline?”

Once ePRO has been judged to be suitable, working collaboratively is key to making sure that the trial runs smoothly. “Work with your sites and sample your sites,” says Crawford. “Does it work for the site? If I’m going to go to a general hospital and they have 28 different ePROs sitting in the study coordinator’s office, is it really feasible? Is there a way we can access some efficiency? A lot of times sites are just forgotten and we just ask them to do what we say.”

Only a matter of time

Regardless of the resistance, the use of ePROs is set to continue to grow over the next few years, which will be accompanied by improvements in the technology. “I think that things will go almost to 100% ePRO,” says Crawford. “What will change is that there will be even more interaction. The algorithms will get better, video will get better so patients can actually record themselves and go to a hub, log in and talk right there either to a healthcare professional or maybe some kind of artificial intelligence.”

There is particular interest in the use of computerised adaptive testing (CAT). A CATenabled ePRO system iteratively learns about the test taker and selects the best item from a large bank of items.

“This tailored questionnaire is often much shorter, reducing patient burden and improving compliance, and can be much more precise, potentially increasing the power of clinical studies,” says Kyte. “Large scale CAT-enabled systems are available, such as PROMIS in the US and the EORTC CAT in Europe.”

Although these developments are exciting, they will bring new challenges. “There is also a huge security hurdle and privacy issue there,” Crawford admits. “Those will be interesting to navigate as well.”

Best practice guidelines

The systems development life cycle (SDLC) is an outline of the activities, tasks, responsibilities and deliverables required to develop a high-quality, validated electronic data collection system. The SDLC methodology employed by the ePRO provider should be scrutinised to determine if it includes, at a minimum, these critical elements:

1. system requirements
2. system design
3. coding/tailoring/software development
4. testing by system provider
5. traceability
6. user acceptance test
7. installation/configuration management
8. decommissioning plan.
Source: The Professional Society for Health Economics and Outcomes Research