            Subj:Harmonization proposal addressing a RIM ballot comment 
            Date:5/15/2003 1:29:14 PM Pacific Daylight Time
            From:gunther@aurora.regenstrief.org
            To:mnm@lists.hl7.org, Scott.M.Robertson@kp.org, 
            hugh_glover@bluewaveinformatics.co.uk, medsig@lists.hl7.org, 
            woody@beelers.com
            Received from Internet: 



Hi,

this is a small proposal to change definition of the 
SubstanceAdministation.rateQty attribute. It was discussed with
Scott Robertson and Hugh Glover after the meeting. This addresses a
negative ballot item on the RIM.


OLD DEFINITION:

Identifies the speed with which the substance is introduced into the 
subject. It is expressed as a period of time over which the quantity of 
the substance specified in doseQuantity is to be administered.

Discussion: This is appropriate for continuously divisible dose forms 
(e.g., liquids, gases). If specified as an interval, the rate should be 
anywhere in the specified range. The 
SubstanceAdministration.rateQuantity is specified as a physical quantity 
in time (a duration). Hence, the rateQuantity is really the denominator 
of the dose rate (the doseQuantity is the numerator). For example, if a 
Ringer's solution is to be given at 100 mL/hour i.v., the doseQuantity 
would be 100 mL and the rateQuantity would be 1 h. Note that there is no 
difference in the actual values of doseQuantity and rateQuantity as long 
as the quotient of both has the same value. In this example, we could 
just as well specify doseQuantity as 50 mL and rateQuantity as 30 min, 
or 200 mL and 2 h or any other combination where the quotient equals 100 
mL/h.

Note that in principle one could again suffice with just the 
doseQuantity attribute specifying the rate in that one attribute (e.g., 
doseQuantity = 100 mL/h). However this practice is not allowed. Systems 
that implement the semantics of units according to the Unified Code for 
Units of Measure would have no problem noting the fact that a 
doseQuantity is really a rate. Other systems however will have 
difficulties to tell an at-once dose from a dose rate from just looking 
at the units. If a system wishes to deal only with a single quantity 
describing the dosage, it can always calculate such a quantity as 
'realDoseQuantity' = doseQuantity * Ingredient.quantity / rateQuantity.

Constraints: rateQuantity.units must be convertible to seconds. I.e. 
rateQuantity must be expressed as a period of time.



NEW DEFINITION

Definition:Identifies the speed with which the substance is introduced 
into the subject. Expressed as a physical (extensive) quantity over 
elapsed time (e.g., examples are 100 mL/h, 1 g/d, 40 mmol/h, etc.)

Discussion: This is appropriate for continuously divisible dose forms 
(e.g., liquids, gases). If specified as an interval, the rate should be 
anywhere in the specified range. This attribute is specified as a 
extensive physical quantity over elapsed time, i.e., a quantity that
could be used for the doseQuantity divided by a duration quantity.






IN ADDITION to this we need a few other changes in explanatory text
of SubstanceAdministration.doseCheckQuantity:

OLD DEFINITION

This attribute identifies the expected quantity to be consumed over a 
period of time. It is used as a verification check on the values 
specified for other values.

Discussion: This attribute should not generally be used; it is only 
provided for a special purpose. In some countries, especially Japan, 
there is a regulatory requirement to note the total daily dose on the 
prescription and associated documentation. The purpose of this 
requirement obviously is to encourage and facilitate reviewing the total 
dose prescribed to avoid over- (or under-) dosage.

Examples:

With Erythromycin 250 mg 1 tablet 3 times a day one can calculate the 
total daily dose as "doseCheckQuantity = doseQuantity (1) * 
Ingredient.quantity (250 mg) * effectiveTime (3 /d) = 750 mg/1d."

For an intravenous example, this term would be "doseCheckQuantity = 
doseQuantity (100 ml) * Ingredient.quantity (5mg/L) / rateQuantity (1 h) 
= 0.5 mg/h" which can be calculated on a daily basis as 
"doseCheckQuantity = 0.5 mg/h * 24 h/d = 12 mg/d."

Rationale: Rather than defining a "total daily dose" attribute as HL7 
v2.3 did, we define this general purpose doseCheckQuantity attribute of 
the Ratio (RTO) data type.

Constraints: invariant(SubstanceAdministration med, RTO max) where 
med.doseCheckQuantity.contains(max) 
{max.numerator.compares(med.doseQuantity); max.denominator.compares(1 
s);} Numerator must be in units comparable to doseQuantity and 
denominator must be a measurement of time.



NEW DEFINITION

This attribute identifies the expected quantity to be consumed over a 
period of time. It is used as a verification check on the values 
specified for other values.

Discussion: This attribute should not generally be used; it is only 
provided for a special purpose. In some countries, especially Japan, 
there is a regulatory requirement to note the total daily dose on the 
prescription and associated documentation. The purpose of this 
requirement obviously is to encourage and facilitate reviewing the total 
dose prescribed to avoid over- (or under-) dosage.

Examples:

With Erythromycin 250 mg 1 tablet 3 times a day one can calculate the 
total daily dose as "doseCheckQuantity = doseQuantity (1) * 
Ingredient.quantity (250 mg) * effectiveTime (3 /d) = 750 mg/1d."

For an intravenous example, where 5 mg/L of ingredient is administered 
at 100 mL/h continuously, this term would be "doseCheckQuantity = 
Ingredient.quantity (5mg/L) * rateQuantity (100 mL/h) = 0.5 mg/h" which 
can be calculated on a daily basis as "doseCheckQuantity = 0.5 mg/h * 24 
h/d = 12 mg/d."

Rationale: Rather than defining a "total daily dose" attribute as HL7 
v2.3 did, we define this general purpose doseCheckQuantity attribute of 
the Ratio (RTO) data type.

Constraints: Numerator must be in units comparable to doseQuantity or
rateQuantity * elapsed time, and denominator must be a measurement of
time.





Rationale: in an attempt to "simplify" the data, the old definition of 
rateQuantity, was too clever to the point where it wouldn't work in 
cases where only the rate is known but no doseQuantity. This new 
definition is more straight-forward.

regards
-Gunther




-- 
Gunther Schadow, M.D., Ph.D.                    gschadow@regenstrief.org
Medical Information Scientist      Regenstrief Institute for Health Care
Adjunct Assistant Professor        Indiana University School of Medicine
tel:1(317)630-7960                         http://aurora.regenstrief.org



---
To access the Mailing List archives, go to: 
http://lists.hl7.org/lyris.pl?enter=mnm


7 in favor passed.