I am searching for a suitable POC device to screen for dyslipidemia among selected adults in a town in Nigeria and the (cheapest and readily available) one I found (Mission 3-in-1 by ACON Labs, San Diego) isn’t as accurate as the venous blood approach. One thing to note is that a recent paper containing details on the investigation of the diagnostic performance of Mission was published last year and the results were largely disappointing. On the other hand, Mission is US FDA-cleared (Class I), so is this is a good basis to work with the device?
@Dan_Eastwood, in a private conversation, proposed a solution to the accuracy problem in case I still decide to use the device but i will like recommendations and ideas from others on the forum, especially health professionals. Thanks.
Briefly, some of the cheaper devices are biased relative to the more expensive (and accurate) models. Kurstjens (2021) give regression equations predicting the bias, and these could be used to apply a “correction” and allow the biased test to be interpreted on the same scale as the more accurate device. The cost is increased variance of the corrected values, but that is small relative to the bias for Mission 3-in-1.
A simpler approach would be to use the regression equation to create an adjusted scale for interpreting the biased device, but requires additional explanation.