Currently submitted to: Journal of Medical Internet Research
Date Submitted: Sep 22, 2020
Open Peer Review Period: Sep 22, 2020 - Nov 17, 2020
(currently open for review)
Towards a universal definition of disease activity score thresholds: The AS135 score
Many study groups have developed scores reflecting disease activity. The result of this fragmentation is a multitude of disease activity scores, even for a single disease. We recently suggested a standardization for the cut-offs. Our standardized system produces a similar wide range of values and facilitates the task of interpreting activity scores for various diseases. However, the formulae used in this article were not perfect as limited to a linear transformation, and without a cell-phone application, the possibility of use by clinicians was low.
To identify and standardize disease activity scores in rheumatology.
We conducted a literature review on disease activity criteria using both a manual approach and in-house computer software (BIBOT) that applies natural language processing (a machine learning artificial intelligence technique) to automatically identify and interpret important words in abstracts published in English between 1.1.1975 and 31.12.2018. Within all extracted disease activity scores, we selected those with cut-off values divided into four classes (remission, low, moderate and high disease activity). We used a linear interpolation to map all these disease activity scores to our new score, the AS135, and developed a smart-phone application to perform the conversion automatically.
A total of 108 activity criteria from various fields (rheumatology, dermatology, gastroenterology, psychiatry, neurology and pneumology) were identified, but it is in rheumatology that we found the most separation into four classes. We built the AS135 score modification for each selected score using a linear interpolation of the existing criteria. It was defined on the interval [0,10], and values 1, 3 and 5 were used as thresholds. These arbitrary thresholds were then associated with the thresholds of the existing criteria, and an interpolation can be calculated, allowing the conversion of the existing criteria into the AS135 criterion. Finally, we created a mobile application that allows each user to obtain both the original value of the activity criteria and the new AS135 value. The use of a linear model to approximate the distribution of each score could be a limitation, but selected scores have been constructed to be interpreted as a linear scale, which makes the approximation performed by AS135 very acceptable.
We developed an application for clinicians that enables the use of a single disease activity score for different inflammatory rheumatic diseases using an intuitive scale, the AS-135 score. Clinical Trial: Not applicable
Request queued. Please wait while the file is being generated. It may take some time.
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.