|
|
| (27 intermediate revisions by 2 users not shown) |
| Line 1: |
Line 1: |
| Patients may show up in our database repeatedly either during the same hospital stay or at a later time. If they show up at a later time, but soon after the last hospital stay, they may qualify as "readmission", which we report in '''<in which, monthly, quarterly, annual or all?>''' | | Patients may show up in our database repeatedly. Our [[L_Log]] records are patient-ward-stay based. Once the [[data processor]] completes the [[centralized data Vetting Process]] (i.e. [[PHIN field|PHINs]] are checked and PseudoPHINs are generated) the PHIN field can be used for linking L_Log records for the same patient. This combining data is stored in [[L Person table]] by [[Encounter processing]]. |
|
| |
|
| == Multiple Encounters ==
| | If a patient is never discharged in between L_Log entries it may be a '''[[Continuous Stay]]'''. |
| <details>
| |
| What is the cut-off for continuous stay detection when moving between ICUs and/or wards? E.g. if patient is discharged from one ward at 0800, how late can he arrive at the next ward and still be considered a continuous stay?
| |
|
| |
|
| What are the fields that have to match to consider two encounters the same patient?
| | If a patient is discharged and then admitted again at a later time, but soon after the last hospital stay, they may qualify as '''[[Readmission]]'''. |
|
| |
|
| == Continuous Hospital Stay == | | == Multiple encounter longitudinal consistency checks == |
| The [[statistician]] uses data from [[Admit From & Discharged To]], [[Med Var 1 - Admit-from Ward]] and [[Med Var 2 - Discharge-to Ward]] to
| | For patients encountered repeatedly some data should be consistent between stays. |
| * link profiles from one ward to another when both wards are within the database program
| |
| * detect a likely continuous stay when a pt is discharged from one of our wards to a non-collected ward and then re-admitted to one of our wards from the same non-collected ward.
| |
|
| |
|
| == Readmissions ==
| | The [[data processor]] runs a set of checks during vetting that attempts to find and make consistent the records for patients encountered previously. |
| For our purposes, readmissions are admissions of patients who were recorded by our database recently.
| |
|
| |
|
| === ICU Definition ===
| | {{Data Integrity Check List}} |
| For ICU, a readmission is a patient where
| |
| * (admit date/time) - (most recent ICU discharge date/time) <= 72 hours
| |
| * is '''not''' admitted for planned and scheduled surgery
| |
| * might be readmitted from Ward or outside hospital
| |
| ====Discussion====
| |
| * why are we considering the most recent ICU discharge date/time only?[[User:Ttenbergen|Ttenbergen]] 11:36, 28 May 2008 (CDT)
| |
|
| |
|
| === Medicine Definition === | | == Related Articles == |
| For medicine, a readmission is a patient where
| | {{Related Articles}} |
| * (admit date/time) - (most recent discharge date/time) <= 7 days after their most recent discharge time
| |
| * is admitted from outside hospital
| |
| * (Legacy: planned surgery used to be excluded, but this stopped because of insufficient data when we stopped collecting some APACHE elements in the medicine program '''<when?>''')
| |
| ====Discussion====
| |
| * is the detection of med readmits for pts discharged from med previously only, or does it include pts who went home directly from ICU?
| |
|
| |
|
|
| |
| {{stub}}
| |
| [[Category:Statistical Analysis]] | | [[Category:Statistical Analysis]] |
| [[Category: Questions_General_Collection]] | | [[Category:Multiple Encounter | * ]] |