Lindus Health interviewing Paul Wicks, Independent Consultant at Wicks Digital Health
Some of the best work I've seen has come out of the COVID-19 trials like RECOVERY and HEAL-COVID. Because these are large scale trials that aim to test a series of different treatments rapidly, they've shown that you can move fastest by being really transparent with everyone involved. That means that whether you're a doctor recruiting for the trial or a patient who's considering enrolling, you can see all the information you need on clear, well-designed websites that feature video interviews with the investigators, plain language summaries of what's involved (in multiple languages), and if you're really a trials nerd, the entire protocol. Both websites also show in real-time how many patients are enrolled in the study and at how many sites - these are features I wish all studies would adopt.
The worst things? Ooh, that's a longer list I'm afraid. Wordy, jargony, incomprehensible recruitment letters and informed consent documents that you'd need a few PhDs to understand are a personal bugbear of mine. A lack of consideration that trial participants are people with lives and responsibilities - lack of support to pick convenient appointments, to have part of the study conducted remotely, slow and clunky systems to get travel expenses reimbursed, and even an explanation of how to take a holiday while being enrolled in a trial are all fairly common. One hot button issue for me is the failure to properly inform participants of the results of a study before the press is informed - people who have altruistically donated their time, hope, and energy should not be finding out that their experimental treatment doesn't work, or that that device implanted in their body won't improve their health, because they've read a press release. Finally, it boggles the mind that when a manufacturer has invested so many resources in running a trial, the published results aren't available "open access" for the whole world to see. These are for the most part simple fixes that would make a big difference to a lot of participants.
The Digital Therapeutics Alliance says "Digital therapeutics (DTx) deliver medical interventions directly to patients using evidence-based, clinically evaluated software to treat, manage, and prevent a broad spectrum of diseases and disorders." For example, these might take the form of smartphone apps that deliver forms of cognitive-behavioural therapy for mental health conditions such as anxiety or depression. What's new about that? Well, we've had a wide array of "health apps" for over a decade now, but hundreds of thousands of them on app stores hadn't undergone rigorous development, testing, or approval. In recent years there has been a tightening of the rules in the US, UK, and EU, amongst other territories, which has led to this new field. It combines medical research, clinical judgement, user-centred design, quality management, and continuous software improvement to deliver new approaches to treating medical conditions.
This differs from traditional treatments like pills, surgery, or medical devices because they are delivered via software. Once a pill is approved it stays the same (pretty much) for the rest of its life, whereas software is always changing and improving, whether to keep up with the latest devices, to fix bugs, or to add new features in response to user feedback. Therefore when we do research on a digital therapeutic it's often more of an iterative process where we go through different stages of proving that the intervention a.) is feasible in theory, b.) is usable and desirable in practice, c.) that it has some effect, d.) that it ideally has a better effect than alternative treatments, e.) that it delivers value in some way so that it's worth paying for, f.) that updates to the software make it better not worse. Because digital therapeutics are so much safer than most drugs and potentially can treat a large population for almost no incremental cost, the potential is great, and we can build in features that allow for greater access such as having digital therapeutics in multiple languages. But bringing new technology to the healthcare market is a slow, hard, frustrating process and it could take a long time to become "business as usual" in the clinic.
With the possible exception of brushing our teeth, humans are not good at sticking to new habits. We let our new year's resolutions lapse by February, we don't take our medication properly, we buy books that we never read, and we download apps that we never use. Trying to understand these patterns of behaviour is really important in digital therapeutics because in many cases we believe there might be some "minimum dose" of say, a sleep hygiene app, or CBT for IBS, or a mindfulness app for nicotine cravings before it can credibly have an effect on people's health. If a digital therapeutic is hard to use, lacks empathy for the user, or if it's just annoying, then people will turn them off or tune them out. A critical part of the development process is understanding what sort of engagement pattern we see from the target population with existing interventions so that we have some benchmark for a new experimental product. A lot of the time the people that develop the apps will hope they've got something as engaging and "sticky" as the best social media, video gaming, or dating apps, but the reality is that something like half of the people who download a new digital health app will never open it, and of those that open it once, half will never open it again. That said, if people get better after only a short course of treatment, we need to ensure we're counting that as a win and not assuming that when people stop engaging it's because they're not getting a benefit.
Patient-centred design work like ethnographic interviews can be really helpful to understand what a patient's day and a week might be like and how their condition affects those around them. In many cases, deciding to take part in a clinical research study will affect spouses, children, parents, babysitters, employers, employees, neighbours, you name it. In my own experience as a caregiver to a person living with a neurological illness, every clinic appointment where I'd take the whole day off work to drive a wheelchair-accessible van down to the hospital for a 170-mile roundtrip would affect an average of 7 people, just to see the doctor for 20 minutes. But just because an intervention is digital, or a trial is decentralized to take part at home, that doesn't mean it's free of burden. People using a smartphone don't necessarily want to complete the 40-item questionnaire they might quite happily have completed on a paper and clipboard in the doctor's waiting room. People working shifts might not want to book a telemedicine visit when they are on tight deadlines and have minimal breaks. But in general, look to the vast array of advice available on the topic of "patient and public involvement" or PPI in research. At a minimum, you should be collecting feedback from your participants before, during, and after your research so you can continually tweak and optimise the study experience. It's not just a good idea anymore, it's a standard people have come to expect.
Speed, focus, and inexperience. Technology teams often lack access to clinical expertise, and they may be unsure how to approach patients and caregivers to ask for advice. A common mistake (with the best of intentions) is to ask healthcare professionals what patients think, want, feel, or need. Well, that's like asking my hairdresser what kind of hat I'd like to wear. I only see her a few times a year, and while she might be a skilled professional with training, good taste, many other customers just like me, and a thorough understanding of the shape of my head, she doesn't really know the answer and I'd be unhappy if I had to abide by her choice. Digital therapeutics companies can also be a little demanding when it comes to asking for my personal data. The first time I use an app I might not feel comfortable entering my date of birth, my postcode, my sexual orientation, or what kind of medication I take. Start by showing your users the potential value of your tool and ensuring they're engaged - after all you can always collect some of that data later on, and there's no point wasting precious minutes of people's scant attention spans filling out demographic questions that means they'll never come back again in the future to experience what else you have to offer. Finally, treat every piece of feedback you receive as if it's come from 10x as many people because many people felt the same thing but didn't have the activation energy to tell you about it.
There's such a contrast between a digital therapeutic with a well designed "onboarding" process and the experience of ticking a thousand boxes in the paper forms of traditional clinical trials. The best digital therapeutics are fun, funny, personalized, helpful, feature modern and accessible interfaces, and connect with other apps and devices in your daily life. The best-case scenario of a traditional trial is often that the research nurse will be friendly and apologise profusely enough for the long wait in hard chairs, the clunky phrasing of the questionnaire, and the lack of information and feedback provided. When the COVID-19 pandemic first struck it suddenly became clear that what was holding us back from using technology wasn't any technical hurdle, but rather the will to change our practice and our habits. When you're sitting in a hospital research site it's just more convenient to bring the patients to you and let them worry about the parking than to have to get the IT department to rig up a stable telemedicine solution or arrange for blood tests to be done via the GP rather than the trusty pathology lab.
In some cases, there is also the thorny issue of reimbursement - if a site is charging a sponsor for staff time and overheads then making the trial experience more efficient is going to give you fewer resources to work with. I think we've just gone through a generational paradigm shift in expectations for how someone taking part in research should be treated. The reality is that people are busy, financially stressed, and impatient - if we don't take this opportunity to catch up to the 21st century then we're going to experience a lot of failed clinical trials in the near future, which would be a tremendous shame. I'm optimistic though that we can learn from everything we've experienced in the past two years and download the collective update patch that we've been putting off for far too long.
Discussing DTx challenges and go-to-market in the UK with Charlotte Lee-Sinclair (ex-Big Health)
A fireside chat, this time by an actual fire! Our conversation with Dr Charlotte Lee-Sinclair is a must-read on DTx challenges and the road to go-to-market in the UK. Read here.
Debunking the 3 biggest myths about decentralised trials
You've probably heard of decentralised clinical trials, but there are a lot of myths out there. Here we debunk the 3 biggest myths.
Product Clinic #6: Releasing, Fast and Slow
Depending on your background, our 6 to 8 week product release cycle will evoke one of two reactions. If you work in the clinical space: Woah, what an incredibly fast release cycle! But if you work in a typical SaaS startup: How quaint, I release 3 times a day! Heard of Daniel Kahnemann's bestseller Thinking, Fast and Slow? Well, there is our inspiration!