Blood Glucose Levels and Dirty Power

The chart shows the effect of dirty power levels on three individuals.  None of these people are diabetic. 

There are 4 sets of measurements:

1. Before eating in a low-level (4 mV) dirty power environment.
2. Thirty minutes after eating in the same low-level dirty power environment.
3. Next day, before eating in a higher-level (10 mV) dirty power environment.
4. Next day, thirty minutes after eating the identical meal from the day before, in a higher-level dirty power environment.

Are there some factors, other than dirty power, giving this apparent correlation?  We will look at Blood Glucose levels Vs Time of Day.

The R2 factor is small when blood glucose is compared to time of day (0.24).  Indeed it is difficult to explain why the correlation factor can be so high when we look at blood glucose Vs dirty power and so low when we look at blood glucose Vs time of day, without concluding that dirty power has a profound effect on blood glucose levels.

When partial correlations are examined to remove any time of day effect, the R2 value is 0.61.  The statistical significance is strong (p < 0.001).

This figure is the same plot as the previous one, except that the Blood Glucose and Dirty Power scales have changed.

This plot is using the exact same set of data, except that the “outlier” is removed. 

This plot is to the same scale as Figure 1 and includes the exact same set of data except the “outlier” is removed.  We see that the R2 factor has been reduced from 0.82 to a still respectable 0.74. 

This is the 1st of a set of graphs that show the effect of dirty power on Dave Stetzer’s blood glucose levels.

While we can see in this scatter plot that the R2 factor is large (0.82), there is a single “outlier” data point.  The “outlier” and a companion data point are shown in pink.  These are 2 measurements, taken 2 hours and 45 minutes apart, in the same house.  One hours and 45 minutes prior to removing filters that substantially reduce the dirty power levels, a bacon cheeseburger and a bowl of vegetable beef soup was consumed. The lower left data point is where filters are installed that removes most of the dirty power.  The upper left is where these same filters have been removed.

This “outlier” data point is real data, but it does raise the question, what would be the correlation factor if the upper right outlier data point were not there?

This picture of dirty power was taken from an oscilloscope in a classroom at Brighten School, Brighten, WI.  It is typical of what is seen all over Wisconsin.

The cause of this dirty power comes from the myriad electrical gadgets and equipment we use.  Recent changes in the technology have resulted in these gadgets and equipment not drawing their power needs continuously, as they did previously, but intermittently at a high frequency.  Home light dimmer switches are but one example.  Such “non-linear” drawing of power is reflected back onto the electrical power system.  The gadgets our neighbors and we use generate such electrical “dirt” as do certain types of variable speed motors.  The “dirt” is also generated by the electrical utilities when they switch their distribution from one circuit to another.