Document Capture (DCC) has had the recent pleasure of collaborating and working with Dr Andy Dickens, an academic research fellow at University of Birmingham. Dr Dickens is working on a large study looking at chronic disease, and is collecting data from patients over a period of time.
The study generates a large quantity of data that, prior to DCC involvement, was manually entered onto a database. The DCC solution did not change the way the study was run; rather, it improved efficiencies of the processes involved and improved data accuracy. Here we take you through the journey…
Study data requirements
Data collection for the study consisted of an assessment visit, which recorded physiological data such as research patients’ height and weight. This assessment, performed at the GP surgery, also included a blowing test to evaluate patients’ lung function. Thereafter there were comprehensive questionnaires administered every 6 months to understand and assess how the patients’ health status changed and developed over the course of time.
The primary challenge was the length of time and resources it would take to manually enter study data. Dr Dickens describes thus:
“For a long time we manually entered data onto a database that we have here at the university. However, to cope with the high volume of data it required several members of staff to be allocated to data entry full-time. That, as you can imagine, is a fairly inefficient use of staff time when the staff can be better utilised doing things to help run the study”
Even with dedicated data entry staff, the team sometimes had large amounts of data backlog. A study assessment took between 90-120 minutes and staggeringly, manually entering the data associated with one assessment took nearly the same time!
The study team implemented policies to standardise data entry and minimise error rates, but a certain degree of keystroke errors were inevitable. Conducting data checks was a substantial undertaking, even on a random sample, considering the large volume of data entered.
It had become clear to Dr Dickens that they could not continue to manually enter data and therefore DCC were brought on board to improve efficiency and increase data accuracy. The return on investment (ROI) is what makes any investment worthwhile. Dr Dickens felt the after sales care and flexibility of DCC was the winner over other providers.
“We wanted a solution that was going to help us work more efficiently. We had an initial meeting with DCC where they put forward what they could offer and the functionality. What sold DCC to us over others and what appealed to me was the flexibility of the DCC system; it worked better for the type of data we were collecting and the had a track record of after sales care. The overall cost of the system, compared to other solutions, was also more competitive.”
How data challenges were solved
Parimal Patel and Nayan Patel came to the University to deliver a one-day training course, giving an introduction to all aspects of the Teleform software. Dr Dickens then set to work developing multiple questionnaires that patients could complete, before being scanned by the research team using Teleform. Dr Dickens explained the support provided to him by Parimal during the development stage:
“Whenever I got into difficulties or was unsure of the best way to design our forms, I contacted DCC for advice and it would often lead to an online session so that Parimal could take me through the process. Parimal was happy for me to contact him whenever necessary and gave me as much time as I needed. The level of support was absolutely invaluable”
DCC set up a fully automated intelligent data capture system, which allowed the research team to scan, store and easily retrieve data from questionnaires. The automated data entry system saved University of Birmingham time and money and thus made the study run more efficiently. Dr Dickens explains the time savings using an example:
“If there is a query with the data, for example, an incorrect value, what I’d have to do is work out which patient it was related to and what practice they were from. Then I would have to go and look in one of our 30-40 plus filing cabinets to find the right patient file, pull it out, pull out the patient copy and find out what the right value is. Whereas with the scanned PDFs all I have to do is go to the folder where the data is being exported to, search for the ID number and retrieve the PDF. This is something I could do from my own desk!”
Additionally, it meant the team could deliver the study within budget, as Dr Dickens explained that research teams have to work within fixed budgets that are set before studies commence.
“We can now get on with the job we are meant to be doing, whereas before we had to pull people off core tasks for the sake of getting data entered. If we continued using manual data entry, we’d be going over budget”
Continuing with manual data entry and thus spending a higher than anticipated portion of the budget on administration would result in having a shorter follow up period for patients, as there would be less funds available for research assistants who perform patient assessments.
Lastly, and perhaps most importantly, the scanning solution meant data can be seen in real time, which means trends can be monitored and acted upon instantaneously:
“One of the things that is very useful that I can look at very quickly from the scanned data things like how many DNAs we had last week, the number of assessments completed and the number of adverse events. I can see that immediately, in real time and this is very useful. For example, I can see for instance, one week we have had a spike in the number of DNAs, I can try and understand whether there is an obvious reason for that or I can work around it and put things in place to improve it. The monitoring aspects has been very good”
Dr Dickens believes that many other academic research teams would benefit from the far reaching impacts of an intelligent data capture system and scanning solution.
To put it into context, a single assessment form (CRF) for this study is 19 pages. Before the team used DCC they could enter approximately four CRFs per hour, whereas they now enter approximately 25 CRFs in the same amount of time. An automated system did not only bring a time saving, it also increased confidence in data accuracy.
As for the value of DCC:
“The after sales was vital, the willingness of the people at your (DCC) end to bend over backwards and help was truly amazing”