eCOA in eSource and early phase trials
Trial without Catheter (TWoC). Plotted by yours truly at bedside.
Introduction
Apologies in advance, as I am writing this essay from a position of inexperience and ignorance. My eHealth experience is heavily biased to early phase / clinical pharmacology trials. Rating scales are used quite often and participant diaries are used occasionally. Where I used to see these, they were still pen and paper based or using third party electronic tools from the big vendors and supplied via the sponsor. They were never integrated with or part of the site’s eSource platform.
Having said that I wanted to sketch out some design ideas that have been influenced and informed by my own recent unexpected visits to urgent care in the US and A&E (accident and emergency) in the UK, and multiple follow-up visits since.
Background
On a recent trip to the US I found myself in a position where I was not able to pee.
When this condition lasts overnight, I can assure you that it becomes quite uncomfortable!
Anyway, I had some ‘emergency plumbing’ fitted and immediately voided 1500 mL of retained urine. Despite the discomfort of the plumbing, that was a blessed relief. I spent the rest of my trip peeing into a bag strapped to my leg - TMI?
On my return to the UK I had a chat with a nice urologist and he did unmentionable things to me without even buying me dinner first.
The good news - not cancer.
The bad news - super enlarged prostate - you are likely to need a re-bore (my term, not his).
In the mean time, let’s get that plumbing removed and have you do a ‘trial without catheter’ (TWoC).
Sweet.
After a day of peeing into pots, recording volumes and the amount of urine retained in the bladder as measured by with an ultrasound scanner, the incredible nurse trained me in how to insert my own temporary plumbing and let me go home.
I must be a terrible student, as I fluffed it twice late at night and could not pass urine. I don’t remember the last time I cried (except from most Friday movie nights). I was not having a good time.
Cue a trip to A&E, the insertion of more long term plumbing (the American one was better!) and an overnight stay.
So what’s all this got to do with eCOA?
Record Keeping
In my interaction with and under the excellent care of the healthcare professionals for my TWoC, I was surprised / not surprised to observe poor compliance with ALCOA principles - measurements were only roughly timed and rarely ‘signed for’ in the paper chart / notes. Printouts from scans were collected in the chart. There was no integration of devices with EHR / EMR systems and in discussion with the staff, no indication of data being entered into computer systems by hand.
Overall, my experience was hugely positive and I feel we reached the correct clinical outcomes (IMHO), but I wondered to myself whether the observations would pass muster and be admissible for a submission to a regulatory authority in support of a marketing authorisation.
Patient Diary
Using the ‘Day One’ journaling app to ‘hack’ together my own personal patient diary.
Here we go. I am getting there. Honest.
I had one follow-up call by the (wonderful, wonderful, I can’t say it enough, wonderful nursing team) about ‘How am I doing?’ I have since done a flow test and I am waiting to see my specialist next week to discuss treatment (surgery) options.
I felt I really wanted a way to connect with the nursing team and to monitor my own progress without calling them every day. I set up my own tracker using some journaling software on my iPhone and iPad to track how often, when and how much I was peeing.
It was very unstructured and subjective. I guess I could have built a Google form or similar and bought myself a measuring jug.
I was pretty sloppy in my own record keeping, regularly forgetting to record when I had a pee.
It also worries me that I am looking at this from my own special viewpoint. Does the clinical team want or need this data when managing patients with well understood conditions and treatment pathways? Don’t they have enough to do already? Do they really want to be hassled by patients filling in online forms all day long and ‘messaging’ them? What are they going to do with all this data?
Design Ideas
Having said all that I pulled together some thoughts for an ancillary ‘eCOA’ module that might compliment an existing eSource or EDC platform.
Importantly, like with most eSource features in early phase trials, the site should have control over the design, build, testing and release of eCOA forms.
Identified Need
Ability to capture structured and unstructured data from patients and to record this in accordance with GxP ALCOA/ALCOA+ principles.
What would this look like?
Scheduled and ad hoc assessments, recording of events
How are you feeling? [Scale] [Ability for clinical staff to convert this or link this to an AE]
Please rate your e.g. pain from 1-10
Did you take anything for the pain? What did you take? [Ability for clinical staff to convert this or link this to a CM]
Record the volume each time you void urine
Record each time you use a catheter to pass urine
Record the amount and type each time you take a drink
Do you want the clinical team to contact you for a non urgent discussion?
Do you want to flag an alert to the clinical team? [Note: the clinical team only monitors these messages in office hours. For urgent support please call {site defined emergency number} or your local urgent care provider]
Anything else you want to add?
Chat with / message the clinical team
Built off CDISC ODM standard
Responsive web design first (app later, ever?) for phone, tablet and computer
Bring Your Own Device (BYOD)
Standard GxP style audit records
Ability to log location of patients and other users as part of data authentication
Data is secure in transit (and at rest of course)
Secure login for trial participants
MFA
Email
SMS? (not favoured any more)
Remember device for x (configurable) days?
Passwords
Encourage use of easily typed pass phrases rather than highly complex random passwords
[Personally I’d like support for password managers to be able to autofill OTP even just for the sake of testing and demos!]
[Future support for passkeys 😄 for testing and demos, that could just as easily be enabled for patients when the time is right]
Reporting - need anything different here?
Want views to show participants their collected data and trends? Keep them involved and motivated? Patient retention?
Allow patients to export their data, excluding any ‘proprietary’ sponsor info? PDF / XLS
Off the table initially
Copyrighted rating scales (or make it clear that it is up to the user to address any usage and validation issues here)
Visual Analogue Scales (the 10 cm line one, unless there is already a mobile friendly alternative)
Crazy meanderings on patient logins / passwords
‘Advanced’ very nice to have option to alert user that their email address has been involved in a breach per https://haveibeenpwned.com. Downside is that this might cause non technical users to panic, and ask ‘what do I do now?’. Maybe flag compromised email accounts to admins? They then have the job of deciding whether they need to do anything. Nothing they can do? Maybe just talked myself out of this feature!
Option to provide feedback to patients that their chosen password has been exposed in a breach via https://haveibeenpwned.com/Passwords - please choose another one. [Or keep a view, list, table, report for admins? Not sure how this would work in practice as passwords should be salted / hashed / one way. Also telling an external (albeit reputable) service what password you are going to use with a site, seems counterproductive? Might have talked my way out of this too!
Geolocation
Is it done routinely already? Been lazy and not done any research!
Why would you do it?
To try to provide additional assurance about the validity of data collected in trials
To be used as a possible signal to detect fraud
How reliable and dependable is it?
Easily ‘fooled’ (deliberately or unintentionally) by use of a VPN?
Throws into question the benefit
System setting to enable this for participant users
Off by default
When on, defaults to on for all trials and all sites in a trial but could be disabled per trial and / or site
System setting to enable this for staff users
Off by default
When on, defaults to on for all trials and all sites in a trial but could be disabled per trial and / or site
Conclusion
An essay where I share too much personal information?
A demonstration of my naïvety and lack of understanding of the current state of play in the eCOA space?
Any useful nuggets that can be carried forward in future eCOA designs, especially for early phase trials?
Feel free to tell me (politely!) where I am going wrong.
