Venous blood is simply venous blood. But capillary blood from fingertip samples contains variable amounts of interstitial fluid, and cell lysis is another issue

Venous blood is simply venous blood. But capillary blood from fingertip samples contains variable amounts of interstitial fluid, and cell lysis is another issue there. There are tests where it doesn't matter - consider the blood glucose - the variability of capillary samples vs venous samples is well understood and the introduced error sufficiently small compared to reference range. But Theranos also tried electrolyte panels, and there you have a problem because the concentrations of ions in blood plasma are very different from the concentrations of ions inside cells. Both are tightly controlled in the healthy body. Slightly less than 50 % of blood is cellular matter, and even small amounts of lysis are sufficient to distort the numbers to render them clinically meaningless. You inevitably have lysis in capillary samples. Carreyrou talks about that in his book.