For the avoidance of doubt - there was no test for synthetic insulin that’s the point.
The test was an immunoassay that tests antibodies to insulin as a proxy for the amount of insulin in the system. It has a reasonably wide margin of error due to cross-reactivity with other substances in body and when used in the lab can result in false positives.
Then there is the issue that infant glucose-insulin metabolism differs significantly from adults. Preterm neonates’ intricate background physiology that maintains glucose in a narrow range in adults does not work in neonates.
The inference that exogenous insulin could be present is from the discrepancy between the apparent insulin level and the c peptide level.
Insulin and c peptide levels are influenced by the organs that clear them. C peptide is produced in equimolar quantities with insulin, and is principally cleared by the kidneys. Insulin is cleared by liver and peripheral tissues in a highly variable way, as well as the kidneys.
The physiology of neonates is immature and the variation in blood volume, liver metabolism and renal function may affect the reliability of insulin and c peptide readings. It may produce variable results requiring multiple samples to get a reliable reading.
For expert perspective on other problems with the test please read the Telegraph article linked a few pages back.