Dear all:

This is a long overdue response to a discussion that occurred months back.  But I really 
can't help commenting on the misunderstandings of measurement error.  In my 
discussion of this, I am not addressing the fact that each of these indicators have a 
different MEANING, but merely the implications from a measurement perspective.

During the prior discussion of using the creamatocrit I read a comment that made a "false 
equivalency" analogy between the "dependability" of pre and post feeding weights and fat 
content measured through creamatocrit.  The claim was that the creamatocrit was as 
"undependable" as pre and post feeding weight checks.  The term "undependability" was 
not used in the post, this is the term I am applying to the discussion because this is the 
actual term for the type of measurement error that was being discussed.  

For those of you who may not have read my posts before "undependability" refers to the 
fact that many measurements of physiologic phenomenon will vary over time because 
their are normal fluctuations.  For instance, you may not eat the same amount in the 
morning that you eat in the afternoon.  Or, your height will be lower in the evening after 
you have been walking or sitting all day long than in the morning when you wake up.  The 
measurements are "accurate" for that moment in time, but don't capture the variation 
that occurs over a span of time.

The measurement of weight with a scale that is ACCURATE to 0.1 oz or 2 g means that it 
is ACCURATE to 0.1 oz or 2 g.  These scales take a triple average to compensate for the 
wiggle factor.   Averages of repeat measurements improve accuracy.  Ditto for 
creamatocrits. They are extremely accurate in determining the fat content OF THAT 
SAMPLE.

The issue is the meaning of these measurements and undependability and this is where 
the comparison between test weighing and creamatocrits falls apart.  

Pre and post feeding weight checks are accurate representations of that SINGLE FEEDING. 
You can accurately assess a SINGLE feed.  The undependability is how that SINGLE feed 
compares to other feedings.  And in this regard, NO SINGLE observation, whether it is 
observing swallows, observing behavior, or diapers is as accurate in determining the 
intake at that SINGLE feed.  And there is plenty of evidence based documentation to 
support this and a lot of speculation that really doesn't come close to refuting the 
EVIDENCE.  HOWEVER, you canNOT judge a baby's feeding behavior from one feed alone. 
And if you only see a mother and baby one time, especially if you only have a short time 
for the assessment you MUST collect the other information on typical swallowing patterns 
and typical behavior to put the feeding in context.  As I wrote before, intake measured by 
a pre and post feeding weight check is not SUFFICIENT to judge how a baby feeds.  It 
matters how hungry the baby was, whether the baby was nudged to eat because of the 
timing of the visit, whether the baby ate well before the feeding -- a zillion different 
scenarios come to mind.  

The creamatocrit has the same problem as pre and post feeding weight checks in that a 
single sample canNOT tell you what is happening at other feedings.  In addition, there is 
an additional parameter to the assessment.  That parameter is that it is usually measured 
from a small sample of the milk removed.   Regardless of the attempts to standardize 
this based on breast fullness, time taken to remove the milk or any other criteria you 
come up with -- there will be huge individual variability from mother to mother as we 
well know from research.  This is yet an additional parameter of "undependability" 
beyond that for any other indicator of a baby's intake at a particular feed.  The 
creamatocrit measures a SAMPLE of a SINGLE feeding and if you took the SAMPLE slightly 
earlier or later in the process it would make a difference to your interpretation.  

So, I would put the comparison the other way around.  The creamatocrit is far less 
dependable than pre and post feeding weight checks.  

Best, Susan Burger

             ***********************************************

Archives: http://community.lsoft.com/archives/LACTNET.html
To reach list owners: [log in to unmask]
Mail all list management commands to: [log in to unmask]
COMMANDS:
1. To temporarily stop your subscription write in the body of an email: set lactnet nomail
2. To start it again: set lactnet mail
3. To unsubscribe: unsubscribe lactnet
4. To get a comprehensive list of rules and directions: get lactnet welcome