groups » Data Management and Statistics » Mobile devices for data capture (EDC) - where's the research?

As part of a dissertation, I'm investigating the accuracy with which data can be captured using mobile devices, primarily targeted at mobile phones in developing regions.

But there seems to be very little research into the accuracy of data capture which surprises me.

- Haller (http://www.ncbi.nlm.nih.gov/pubmed/19567799) says 'Despite the increasing use of handheld computers for electronic data collection in clinical research, these devices should be used with caution. They double the duration of the data entry process and significantly increase the risk of typing errors and missing data.'
- And Shelby-James (http://www.trialsjournal.com/content/8/1/5/) states '...unacceptable high error rates occurred with hand held computers...'

My hypothesis (based on what I've seen of various apps) is that the User Interface is too 'computer-like' to be a good fit with the generic mobile phone handset. If I can find existing research in the field and my hypothesis proves out, it will make my case stronger.

But despite the plethora of data capture apps out there (with Google, I've found 35 of which 11 are specific to clinical trials ), I can find hardly any research ;-(

IS there anyone out there who can point me to research on using any mobile devices (tablet, PDA, smart-phone or ordinary mobile phone)?

Reply

data capture EDC handset mobile PDA phone

  • mikewor Mike Workman - Senior Contributor 2 Aug 2012

    Thanks very much Cdiaz. It was Walther's paper that really stimulated me to chase my ideas. And I'd already seen Grayson's document (which I class as advertising, not research). At least I know that professionals in the health trials arena have as much difficulty as I did in finding pertinent research. I'll certainly let you know the results, and maybe I can get it published next year.
    Thanks once again
    Mike

  • cdiaz cdiaz 27 Jul 2012

    Hello Mike

    you are absolutely right, it is very hard to find data on mobile phones specifically. I found this paper in plosone which is interesing, and compares different types of electronic device: http://www.plosone.org/article/info:doi/10.1371/journal.pone.0025348

    This one has some interesting info on using mobile phones for data capture - though not an evaluation as such: http://www.google.co.uk/url?sa=t&rct=j&q=mobile%20phones%20for%20data%20capture%20in%20clinical%20trials&source=web&cd=7&ved=0CHwQFjAG&url=http%3A%2F%2Fwww.touchbriefings.com%2Fdownload.cfm%3FfileID%3D15694%26action%3DdownloadFile&ei=tVkSUOOBIoW-0QXrjIHwAQ&usg=AFQjCNFPGRiUlQoqMWKbCyUEj26PQgEdLQ

    However, it seems there's a real gap in research on the particular topic you mention. I'll be really interested to hear what you find.

    Perhaps others know of more research in this area?

  • mikewor Mike Workman - Senior Contributor 10 Jul 2012

    Thnaks Cdiaz, I'll eagerly wait for your response. We too have the 'carpark effect', although I call it 'sitting at home' effect. In one study here in South Africa they found a researcher who was seemingly able to walk at 70 km/hr from one interview to the next - they found she was sitting on her couch at home making up the answers!
    Most of what I've found in terms of accuracy is of the 'reduces queries by 70%' type of evaluation. Forgive me for being a sceptic, but if they no longer have a CRF as the 'gold standard', how do they measure the number of queries? There are a few studies on the use of PDAs, but these date back to 2005 and earlier before the explosion of mobile phones. So while they give some insight, they are not directly comparable to what I'm trying to do.
    SO once again thanks for your interest.

  • cdiaz cdiaz 8 Jul 2012

    Dear Mike

    This is a really interesting question! I don't think I can answer very specifically RIGHT now but I am consulting some people and will get back to you. I have previously been involved with a consultancy, and a group there was doing some work into looking at capturing electronic patient data (e.g. giving patients PDAs to respond to questionnaires about their health). This is slightly different I think to what you're asking but the results may be of use to you?

    From memory (and I will try and find some research that backs this up but thought initial feedback might be of use to you!) studies found that capturing patient reported outcomes (PROs) electronically was very much more accurate than using paper questionnaires, and I think this could also apply to the types of situations you mention. The reasons for this are:

    1) electronic data capture forms can force you to fill in fields, so you cannot leave them blank (eliminating the problem of missing data)

    2) Electronic data capture forms can dictate the type of information in a field - e.g. if the result should be a numerical value, an eDCF will only let you enter a numeral, while a paper form could let you write anything.

    3) It would be possible to programme the device to download straight into your study database, meaning that data entry is less arduous (could just mean data checking rather than data entering, for example!).

    4) This is specific to PROs specifically, but may be of interest - I believe there is something called, in the US, the 'carpark effect' - i.e. patients who should have been filling daily questionnaires instead fill them in 'in the carpark' (or waiting room, or wherever) immediately before seeing the doctor rather than doing them daily. An eDCF can record the time and date that the data was entered, and can also be set up to have reminders or alarms reminding the patient to fill them in. Presumably, this could also apply in some trial settings with other types of data capture than just PROs.

    I hope this helps a little. I'll do some reading and get back to you, as this is a very interesting point!

    Thank you

Please Sign in (or Register) to view further.