My home meter always reads about 30 mg less than the doctor’s office. They send the blood out to a lab for the results. I just bought a new meter that reads in plasma results like a lab so I would assume it is correct. The manufacturer said that blood in a test tube will rise 7mg for every hour it is not tested after it is withdrawn from the body. Is this true? If withdrawn at 8:30 and tested in a lab at 12:30 would explain the difference I imagine. Can you give me any information on this?

Your question raises some important issues in relation to blood sugar measurements. First of all, important errors occur when blood is taken into any tube other than one containing fluoride or when it is left uncentrifuged for more than 30 minutes or so. The effect is to show an apparently lower blood sugar. Secondly, it can be a source of error to compare two methodologies in two settings without any statistical assessment of variation in either. You need to run the same sample twenty or more times in each setting and to calculate the 95% confidence limits of a measurement: for glucose this should be less than about 6%. With this information you can judge whether any detected difference could have been achieved by chance. The best way to find out if two methodologies give intrinsically different results is to compare the figures for proficiency testing published by the American College of Clinical Pathology which any commercial laboratory will have.

Since your results seem to always read less, I would guess that the problem is not due to delay in separating cells and serum and that it is more likely that there is an error in your own technic. This latter factor you could check by calibrating your meter against the standard provided and by revising your technique with a nurse educator.

DOB