Defining request triggers

[GitJamis]

[isaacvetter]

Hey @GitJamis ,

This is clearly an important issue. I really like your .

which suggests that a CPOE system could reasonably support:

• [medication]-[select|dose|route|frequency|duration|link|sign]

Most steps for a procedure would be different, but select and sign would remain applicable. Therefore, a more reuseable set of hooks that weren't specific to medication or procedure could be order-select or order-sign.

Two questions:

1. From an academic, clinical perspective, what would the similar set of steps be for a procedure?
2. Any concerns with my proposing order-select and order-signand then simply identifying medication-prescribe as a medication specific subset of order-sign ? (Effectively, implementing Matt's suggestion from make medication-prescribe a generic order hook, update order-* hook names #411).

Isaac

[GitJamis]

Hi Isaac, Thank you for commenting. Please see my responses to the questions below. Re: Two questions: 1. From an academic, clinical perspective, what would the similar set of steps be for a procedure? I am not sure that I am qualified to comment on a procedure, but I believe it would be analogous to a medication order. A procedure would be selected then details entered and accept to complete and/or sign. 1. Any concerns with my proposing order-select and order-sign and then simply identifying medication-prescribe as a medication specific subset of order-sign ? (Effectively, implementing Matt's suggestion from #411<#411>). I have a couple concerns with order-select and -sign. The first concern stems from my lack of knowledge from other types of orders (e.g., procedures). I worry by making the hooks generic across other orders it will lead to growing requirements for each hook with more use-cases. My second concern is having multiple orders per select (e.g., order set) and sign (e.g., multiple medication orders completed but signed once) events. I am unsure how to get around the order-select since the medications will be prepopulated. But I like the idea of a “finalizing” event for a specific medication order rather than specifying “sign”. It seems to reduce the workload vendors have made it easier to sign off on multiple orders with one action. With the PDDI use case we could be providing information for an order that was placed a couple orders prior or even at a different EHR interaction, when the orders are signed. Best, Tom From: Isaac Vetter <[email protected]> Reply-To: cds-hooks/docs <[email protected]> Date: Tuesday, November 13, 2018 at 1:57 PM To: cds-hooks/docs <[email protected]> Cc: Thomas Reese <[email protected]>, Mention <[email protected]> Subject: Re: [cds-hooks/docs] Defining request triggers (#418) Hey @GitJamis<https://github.com/GitJamis> , This is clearly an important issue. I really like your [flow diagram] <https://camo.githubusercontent.com/36d16a847407f264b07a0b6bdcb14f3027606e6b/687474703a2f2f686c372e6f72672f666869722f75762f706464692f323031385365702f6173736574732f696d616765732f43504f455f776f726b666c6f772e737667> . which suggests that a CPOE system could reasonably support: * [medication]-[select|dose|route|frequency|duration|link|sign] Most steps for a procedure would be different, but select and sign would remain applicable. Therefore, a more reuseable set of hooks that weren't specific to medication or procedure could be order-select or order-sign. Two questions: 1. From an academic, clinical perspective, what would the similar set of steps be for a procedure? 2. Any concerns with my proposing order-select and order-signand then simply identifying medication-prescribe as a medication specific subset of order-sign ? (Effectively, implementing Matt's suggestion from #411<#411>). Isaac — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#418 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/ATYnYduVW3NHCR0uanw0Y5ehWBCad8B9ks5uuzI0gaJpZM4YAbWd>.