Selected PHD to PCD Mapping Issues (out of a total of 35)

Paul Schluter, on behalf of the IHE and Continua WAN team 2009-09-24

Using pre-coordinated temperature observations as site locations (10408, Thermometer) [pg 33, PC63] [ACCEPTABLE?]

REFID / Description
MDC_DEV_SPEC_PROFILE_TEMP / Thermometer MDS /  example of dotted REFIDs
to show observation hierarchy
... MDC_TEMP_BODY / Body Temperature
Description / OBX-3 REFID / OBX-6 Units / OBX-20 Site
Thermometer MDS
Body Temperature / DISCUSSION: The codes and REFIDs listed in OBX-20 really belong here as pre-coordinated temperatures+sites (depends on how you interpret the 10408 standard)
150364^MDC_TEMP_BODY^MDC or
188428^MDC_TEMP_EAR^MDC or
188432^MDC_TEMP_FINGER^MDC or
188456^MDC_TEMP_GIT^MDC or
188424^MDC_TEMP_ORAL^MDC or 188420^MDC_TEMP_RECT^MDC or
188448^MDC_TEMP_TOE^MDC or
150392^MDC_TEMP_TYMP^MDC or
188452^MDC_TEMP_AXILLA^MDC / 268192^MDC_DIM_DEGC^MDC or 266560^MDC_DIM_FAHR^MDC / 150364^MDC_TEMP_BODY^MDC or
188428^MDC_TEMP_EAR^MDC or
188432^MDC_TEMP_FINGER^MDC or
188456^MDC_TEMP_GIT^MDC or
188424^MDC_TEMP_ORAL^MDC or
188420^MDC_TEMP_RECT^MDC or
188448^MDC_TEMP_TOE^MDC or
150392^MDC_TEMP_TYMP^MDC or
188452^MDC_TEMP_AXILLA^MDC

Issue: The ‘site’ enumerations in OBX-20 are really pre-coordinated temperature+site terms (that’s how they are used in PCD-01).

Observations that should/could be enumerated values? (10404, Pulse Oximeter) [pg 26, RC36] [ACCEPTABLE?]

Specified Mapping:

OBX-3 REFID / Description / OBX-5 Value
... MDC_PULS_OXIM_PULS_RATE / Pulse Rate / 64
.... MDC_MODALITY_FAST or
.... MDC_MODALITY_SLOW or
.... MDC_MODALITY_SPOT / Pulse Rate Modality
If there are multiple modalities in the Supplemental-Types then there should be an OBX for each one

Issue #1: This should have been an enumerated value, but since -10404 has been balloted and approved, leave ‘as is’?

Essentially, MDC_MODALITY_FAST has an implied boolean value of ‘true’ in the example above.

Alternative:

OBX-3 REFID / Description / OBX-5 Value
... MDC_PULS_OXIM_PULS_RATE / Pulse Rate / 64
.... MDC_PULS_OXIM_MODALITY / Pulse Oximeter Modality / 0^pulse-modality-fast^MDC_BITS_10404
1^pulse-modality-slow^MDC_BITS_10404
2^pulse-modality-spot^MDC_BITS_10404

For now, leave ‘as is’ -- use the upper example. The ‘alternative’, however, is far more consistent with all the other SpO2 measurement status values are reported.

Statistical Extensions for Observations (10441, Cardiovascular Fitness and Activity Monitor) [pg 44, PS82] [MAJOR]

Description / OBX-2 / OBX-3 REFID / OBX-4 / OBX-5 Value
Speed / NM / 8454254^MDC_HF_SPEED^MDC / 1.0.1.11 / 37.3
Measurement Type (Speed) / CWE / 67883^MDC_ATTR_ID_PHYSIO^MDC / 1.0.1.11.1 / 8456144^MDC_HF_MEAN_NULL_EXCLUDE^MDC or
8456145^MDC_HF_MEAN_NULL_INCLUDE^MDC or
8456146^MDC_HF_MAX^MDC or
8456147^MDC_HF_MIN^MDC

Issue #1: _MIN, _MAX and _MEAN discriminators are not defined for many of the 104xx physical quantity observations.

Issue #2: It would be more consistent with HL7 V2 and PCD-01 to define a common set of statistical extension attributes that can be easily be extended to support the 20+ ‘parametric probability distribution’ defined in HL7 V3 as dot-level-5 ‘observation attributes’ that convey the value in OBX-5 (like any OBX should do) rather than using them to ‘modify’ the meaning of the ‘speed’ variable. [Also, they would apply to all specializations, not just the _HF_ specialization.]

8456144^MDC_HF_MEAN_NULL_EXCLUDE^MDC or

8456145^MDC_HF_MEAN_NULL_INCLUDE^MDC or

8456146^MDC_HF_MAX^MDC or

8456147^MDC_HF_MIN^MDC

Description / OBX-2 / OBX-3 REFID / OBX-4 / OBX-5
Speed / NM / 8454254^MDC_HF_SPEED^MDC / 1.0.1.11 / 37.3
Speed - minimum
Speed - maximum / NM
NM / 8456147^MDC_HF_MIN^MDC
8456146^MDC_HF_MAX^MDC / 1.0.1.11.1
1.0.1.11.2 / 36.2
40.5

Issue #3: How would we add the HL7 V3 ‘parametric probability distributions’, which introduces 35 new statistical quantities that could be applied to practically
any numeric observation. These would be best implemented as standard attributes that report the value their OBX-5, rather than ‘directly modifying’ the parent OBX.

ppd_NULL ( mu, sigma )Unknown distribution (null) [mu, sigma]

ppd_U ( a, b )Uniform distribution U [a, b]

ppd_TRI ( a, b, c )Triangular distribution TRI [a, b, c]

ppd_TRP ( a, b, c, d )Normal (Gaussian) distribution N [mu, sigma]

ppd_N ( mu, sigma )Normal (Gaussian) distribution N [mu, sigma]

ppd_LN ( MU, SIGMA, location )Log-Normal distribution LN [MU, SIGMA]

ppd_G ( alpha, beta, location )Gamma distribution G [α, β]

ppd_E ( beta, location )Exponential distribution E [β]

ppd_X2 ( nfree, scale, location )2 (chi-square) distribution X2 []

ppd_T ( nfree, scale, location )t (Student) T []

ppd_F ( nfree1, nfree2,scale, location )F distribution F [1, 2]

ppd_B ( alpha, beta, scale, location )β distribution B [α, β]

General Observations:

1.There has been little participation by IEEE 11073 experts in reviewing the WAN mapping effort, either by the IHE Ambassadors or IEEE 11073 members who are already members of Continua. We need more ‘11073 classically trained’ eyes reviewing the PHD and WAN mapping effort!

2.Tradeoff between (1) simplifying PHD to PCD conversion vs. (2) preserving ‘integrity’ of HL7 V2 OBX ‘observational’ paradigm.

Note: All examples are from the Continua ‘WAN Payload Mapping’ document dated 2009-09-18 by Randy Carroll and other members of the Continua WAN team.

Note: Many of the oddities noted here are due to the conventions adopted by the PHD protocol, and not necessarily the WAN mapping effort!

Text in red reflects discussion during 2009-09-24 Q1 discussion at the IEEE 11073 meeting in Altanta.

PHD2PCD.MappingIssues.1a.doc- 1 -2009-09-24T10