In this study reference ranges were established for autonomic and peri
pheral nerve tests in 122 non-diabetic adolescents. Regression analysi
s was used to evaluate the effect of age and gender on neurological fu
nction. Increasing age was associated with: less heart rate variabilit
y during deep breathing (p = 0.03), higher thermal threshold for cold
at the wrist (p = 0.009), and higher vibration threshold at the toe (p
= 0.001) and medial malleolus (p = 0.01). Male gender was associated
with higher Valsalva ratio (p = 0.0004), higher thermal threshold for
hot at the foot (p = 0.002), and higher vibration threshold at the mal
leolus (p = 0.03). The REFVAL programme was used to determine parametr
ic or non-parametric reference limits: the 5 % limits for autonomic an
d 95 % limits for peripheral tests. One hundred and eighty-one adolesc
ents with diabetes were studied under identical conditions and similar
effects of age and gender were found. Twenty-eight percent of the gro
up with diabetes had at least one abnormal autonomic test result out o
f four (expected 18.5 %); 24 % had at least one abnormal peripheral te
st result out of six (expected 26.5 %). Glycaemic control was associat
ed with autonomic (p = 0.04) but not peripheral abnormalities. Using m
ultiple regression analysis and adjusting for age and gender, there wa
s no effect of diabetes duration or glycaemic control on neurological
function.