Question

Khan, Sobki, and Alhomida (2015) examined 75 patients to assess the association between fasting blood sugar (FBS) measured iln mmol/l and glycosylated hemoglobin (HbA1c) levels measured as a percentage and reported the following regression equation, which can be used to estimate an HbA1c level from a FBS level. The researchers report that ethnicity was not measured in this study but has been reported to be significant in other similar studies. If the study were repeated and ethnicity was included, and if none of the independent variables was then significant, we would know the original study that reported FBS was significantly related to HbA1c had made what type of error?

Answer 1

Let us assume the regression line  HbA1c = 0.387 (FBS) + 4.855.

Blood glucose will be highly variable by increased rapidly after a carbohydrate meal and falls down to the steady fasting state. Fasting state is a dynamic state, body removes glucose from the blood balance by the glycogen breakdown and glyconeogenesis.

As glucose levels are lower when fasting,the average glucose was lower compared to the non fasting samples. As a general rule, every 1% change in A1C associated with 30mg/dl change in estimated average glucose. The glucose bound hemoglobin or HbA1C provide average blood glucose level as it mix with glucose and hemoglobin HbA1C level directly proportional to the blood glucose level. Fasting glucose test measure glucose concentration.

80% minimum sensitivity and specificity of HbA1C in the prediction of FBS > 126mg/dl precision of 3% and 95% of significant level.

Based on this data conventional cutoff point of 6%, each 1unit FBS in the prediction of high HbA1C based on the cutoff points.There is positive result associated with HbA1C and FBS. If HbA1C had greater sensitivity and negative predictor value in detection of FBS.This study highlighted a straight positive result associated between HbA1C and FBS.