There is a lot of published peer literature on the subject of obtaining quality blood samples and on the variability of results. A good starting point for anyone interested in the subject is 'A1C Versus Glucose Testing:A Comparison', which explains the problems:
"Numerous factors may contribute to this lack of reproducibility. These are elaborated below." http://care.diabetesjournals.org/conten ... 8.full.pdf
Whilst that particular study demonstrated variations in results within one sample set under controlled conditions, other studies highlight the differences between blood sample sites and quality of sample:
Potential inaccuracy of finger prick blood samples.
Comparison of glucose levels in capillary blood samples obtained from a variety of body sites.
Rapid changes in postprandial blood glucose produce concentration differences at finger, forearm, and thigh sampling sites.
and many others
Arterial blood samples provide the most accurate data, venous samples the next level with capillary and blood spot the lowest and most variable samples as they contain undetermined proportions of blood from arterioles, venules, capillaries, plus interstitial and intracellular fluids. As you can see from this more 'media friendly' web page on capillary samples from the University Medical Centre,
"Capillary blood sampling may provide inaccurate results, such as falsely elevated sugar, electrolyte, and blood count values." http://www.umm.edu/ency/article/003427.htm
My wife who is a Biomedical Scientist specialising in haematology informs me that they run controls on their equipment many times each day. Most people with home equipment don't even do it once per week so in addition to the quality of the blood sample one has also to work within the constraints of accuracy of equipment. For example, my meter showed a plasma bg count of 8.8 mmol/L before I went for lab testing. When I returned, the reading was 8.0 mmol/L. These are of course capillary tests using a finger pricker. The result of the lab test using venous blood was 7.6 mmol/L. A clear difference between capillary and venous blood results entirely in keeping wth the published literature.