BACKGROUND
The registration of unsupervised Lateral Flow Test results poses concerns and has not yet been addressed, so the reader should bear with me as this manuscript is of a new category. One reliable reporting method currently being employed in an unsupervised scenario involves a medic reading off the remote candidate’s scanned test device. A common variant of this method entails a medic supervising the test in live video. However, these methods are only used because in those cases, there are many medics compared to the number of candidates. They would not be applicable in the adverse scenario where there are too many candidates compared to available medics, such that any form of supervision is infeasible. Again, I emphasise that I have not encountered any previous research on this matter that I can refer to here. However, I came to these observations after carefully exploring the various commercial products on the market
Scanning or video solutions require a high number of available medics to open thousands of emails or be present on video. Imagine a pandemic scenario where urgent mass evacuation and isolation is necessary. A comparable scenario is the familiar back-to-office scheme, which many of us have been a part of. I observed that some candidates simply did not even use the kits, but instead, just reported negative, which was the desired result. This is because it was easy to cheat and there was no mechanism to catch or discourage this.
OBJECTIVE
This problem has not yet been addressed in previous papers, and is the subject of this paper. Here, I suggest a feasible solution to cheating in the registration following an unsupervised test. I also provide a working prototype which is re-usable for future pandemics of this nature.
METHODS
Ad-hoc Mathematical analysis
Computer programming
RESULTS
The computer prototype I created as a complement to this paper and is available on my github repository.
This solution becomes very obvious once the problem is introduced to the keen mind, and I believe that breaking it down any further would result in the paper being unnecessarily too long. The reader can immediately infer the following:
The program shows that a cheating candidate would only have to guess in order to register their fake results
If three groups of test strips are produced and shipped in equal numbers, the cheat would register a valid result at a success rate of at least 0.33. This is as a result of simple probability. This can be tested by running the program
This would discourage cheating, because the cheat would rather perform the test and report feasible results than attempt to cheat and be caught, wasting their own time
This solution should be applicable to any such unsupervised test, so it is re-usable for future pandemics of this nature.
CONCLUSIONS
The way in which unsupervised Rapid Antigen Tests are registered is a major concern. This paper suggests one feasible solution. The computer prototype I created as a complement to this paper and is available on my github repository.
CLINICALTRIAL
https://github.com/Donald-Besong/Covid_Test_Reporting
