Hi, MHD!

I've just spotted a fascinating bit of research,
<http://www.bmj.com/cgi/content/full/bmj.39534.571644.BEv1?rss=1>, which
investigated the effectiveness of blood sugar testing in newly diagnosed
T2s.  The study was led by Dr. Maurice O'Kane of Altnagelvin Hospital,
Western Health and Social Care Trust, Londonderry, Northern Ireland.

There were 184 participants in the study, 111 men and (presumably ;-) 73
women, and they were on average fairly old (~57 years).  96 subjects were
instructed to do blood glucose testing 8 times a week, and advised what
to do about high or low readings.  The other 88 weren't.  For the testing
group, a "compliance level" was defined as "doing at least 80% of the
required tests".  Only 63 of the 96 in the "testing" group managed this
80% threshold.  I think that the study results include the results from
all 96 people, but I'm not quite sure.  All were tested in the same
diabetic clinic at three monthly intervals, the study lasting for a year.

At all of the measurement times in study, there were _no_ statistically
significant differences between the two groups in their hBA1c, occurence
of hypoglycaemia or use of T2 drugs.

However,

   Participants in the self monitoring group were more depressed,
   scoring 6 points higher (that is, 6%) on the depression subscale of
   the well-being questionnaire at 12 months (P=0.01), and there was a
   trend towards increased anxiety.

The article doesn't make clear how big a difference this 6% represents
(it gives referencs to the (offline) descriptions of the psychological
tests), but the authors clearly regard it as big enough to be important.

The authors give their views on the possible mechanisms for the
depression:

   Anecdotal and other evidence suggests that some patients consider
   monitoring uncomfortable, intrusive, and unpleasant.  An interesting
   finding of our study was that monitoring was associated with a 6%
   higher score on a depression subscale and a trend towards increased
   anxiety, although satisfaction with treatment was unchanged.  This
   supports the results of Franciosi et al, who also found higher levels
   of distress, depressive symptoms, and anxiety in patients who self
   monitored, and the qualitative findings of Peel et al.  This possible
   negative effect of monitoring might be important and merits further
   investigation.  Given that glycaemic control rapidly improved to
   satisfactory levels during the study, the negative effect might
   relate less to feelings of powerlessness in the face of high blood
   glucose readings than to the enforced discipline of regular
   monitoring without any tangible gain.  This possibility should be
   considered when patients with a new diagnosis of diabetes are
   introduced to monitoring.

So, what are we to make of this?  That BS testing, of itself, causes
depression and anxiety should surprise nobody, and it's refreshing to see
this figuring prominently in a proper study.

That BS testing 8 times a week brings no benefit is ambiguous: either
(i) BS testing for new T2s is not useful at all; or
(ii) To be useful for new T2s, BS testing must be done more frequently
 than 8 times a week.

It would be interesting to see this study repeated for
(i) T1s (Hey, I'd volunteer ;-);
(ii) more "experienced" T2s;
and, perhaps most significantly,
(iii) testing much more frequently than 8 times a week (say, 4 times a
 day).

My own guess on (iii) is that it would lower the testing group's hBa1c's
to the point where they'd be statistically significantly lower, but that
this difference would nonetheless be small.  I believe that it would also
raise the already high (34%) "non-compliance" figure, and would raise the
depression level, too.