[Previous Months][Date Index][Thread Index][Join - Register][Login]
  [Message Prev][Message Next][Thread Prev][Thread Next]

RE: [IP] Good Planning and Statistics - (was private reply to Jim H.)


Might be better to post this privately.   I am (at this moment) intending 
to post to the list so I take more time to find the "correct" words that 
convey my thoughts.  It appears that the words I used (and/or what you 
read) were not even close to the thoughts I was trying to convey.

Don't all studies include how many subjects completed the research 
period?  Part of this, I would assume, comes from good planning by the 
researchers.  It also is affected by how beneficial the research is to the 
subjects - either directly or for other reasons.

Was the DCCT on the tail of the bell-curve compared to other studies 
related to subject completion (99%?)

Were all medical costs covered for the subjects in the DCCT?  (supplies, 
tests, treatment).   In addition, do I remember that they also had access 
to a very large support team?

During the DCCT the intensive treatment group "maintained" an A1c average 
in the low 7s for an average time of 6.5 years.

After the subjects were "released to the regular health care system" (EDIC) 
the results of the A1c seem (at least to me) to show it is much more 
difficult to maintain "good" A1c values in the "real world".   The long 
term results now are showing increases in A1c.

Diabetes Health May 2000  - Index - Methodist Health Care System

                          At the end of the EDIC study, the average
                          glycosylated hemoglobin values (hemoglobin
                          A1c ) were 8.2 percent for the former
                          conventional therapy group, compared to 7.9
                          percent for the former intensive therapy
                          group (normal hemoglobin A1c is 6 percent).

The following is not current - has standardization of the A1c been acomplished?
EDIC/NGSP Background and Rationale:

The landmark nine-year Diabetes Control and Complications Trial (DCCT), 
completed in 1993, showed that the risk for development and progression of 
the chronic
complications of diabetes is closely related to the degree of glycemic 
control, as measured by glycohemoglobin (GHB) determinations (5). The DCCT 
also provided a large body of data relating GHB values to mean blood 
glucose. Thus, the DCCT results have set the stage for establishing 
specific diabetes treatment goals using GHB as an index of mean blood 
glucose. However, the fact that GHB assay methods have not been 
standardized among laboratories has prevented optimal use of the test. 
Recent studies (6-10) have shown the advantages and feasibility of 
standardizing GHB assays, and a candidate reference method for GHB has been 
proposed (9,12).

At 10:12 AM 09/04/2002 -0400, you wrote:
>email @ redacted wrote:
> > Specifically, related to the DCCT I have heard that 99% of the subjects
> > completed the study.   I also understand the 1% included one or more
> > deaths.....   Any idea how the DCCT compares in subject completion with
> > other studies of similar duration and number of participants.
>It's really not useful to compare studies in this manner.  Each study has
>its own set of assumptions.  Excellent participation and low loss to follow
>up (known as censored data) doesn't mean much in the way of validity of the
>outcomes and conclusions.

I never intended to question the "validity of the outcomes and conclusions".

> > Since the DCCT results were so conclusive (study stopped early), might
> > there have been an ethical situation of continuing the study with a
> > "conventional treatment group"?
>Almost certainly.  The results of the DCCT were so dramatic that tight
>control immediately became standard of care.  Giving less than standard of
>care in a research protocol is unethical.
>I think I have to interject here that a critical part is in research
>protocols.  Many patients still do not exercise tight control.  This may be
>a matter of physician training, or it may be poor compliance by the patient.
>But this is clinical care, not research.  It would be improper for any IRB*
>to approve a protocol that provided for less than standard of care.
>Jim Handsfield

You mention clinical care - Are we attempting to reach "research results" 
using only clinical care.    Have we set a "standard" that few people are 
able to maintain for an extended period of time?

Are we as individuals willing to talk honestly about how our long term 
"control" is?
And I sure hope the "standards" we are trying to reach are at the "center 
of the bell curve"....

realistically attainable?????....

Take what you like and leave the rest,

Jim S.
email @ redacted
for HELP or to subscribe/unsubscribe, contact: HELP@insulin-pumpers.org
send a DONATION http://www.Insulin-Pumpers.org/donate.shtml