FHIR Chat · RiskAssessment decimal · implementers

Stream: implementers

Topic: RiskAssessment decimal


view this post on Zulip Grahame Grieve (Oct 19 2021 at 19:24):

RiskAssessment.prediction.probability[x] can be a decimal or a range. If it's decimal, it's a percentage from 0 - 100. Why is it a decimal? Is it expected that the percentage value has decimal places? e.g. 0.001% - is that significant?

view this post on Zulip Grahame Grieve (Oct 19 2021 at 19:28):

while I'm here, the invariant on RiskAssessment.prediction is

probability is decimal implies (probability as decimal) <= 100

That's not satisfied when probably is null, so it in effect enforces there to be a probably present. I think that's not what was intended? This has come up for an implementer - should we work around that (the validator can override selected invariants)

view this post on Zulip Grahame Grieve (Oct 19 2021 at 19:29):

@Lloyd McKenzie @Mark Iantorno

view this post on Zulip Lloyd McKenzie (Oct 19 2021 at 19:41):

Agree that the intention was not to force probability to be present. I don't know that .001% matters, but 72.3% is certainly possible. Particularly when you're dealing with low number - e.g. 1.2% vs. 1.7%


Last updated: Apr 12 2022 at 19:14 UTC