-- SvetlanaEden - 09 Nov 2010


Population Definitions

Population One

  1. Inpatient Scr greater than 2 (InptCrCtr > 2)
  2. Non-missing baseline Scr (MostRecentPreadmitOutptCr - non-missing)
  3. DaysFirstEncounterPostDC <= 365

Population Two

All patients from Population One plus
  1. Days from admission to death is non-negative (DaysDeathFromAdm>=0)
  2. No preadmission dialysis appointment (PreAdmitDialysisCtrAppt = 0)
  3. PreAdmDialysisCtrICD = 0
  4. Most recent outpatient GFR >=15 (MostRecentOutPtGFR >= 15)
  5. No preadmission renal transplantations (PreAdmit_RenalTransplant == 0 )
  6. No dialysis post peak 23 to 30 days (PostPeak_23_30_Dialysis == 0)
  7. Length of stay is 30 days or less (LOSDays <= 30)
  8. No preadmission renal appointment (PreAdmitRenalClinicScheduledCtr == 0)
  9. LastBefore30DayPostPeakGFR < 60
  10. No preadmission renal visit (PreAdmitRenalClinicSeen == 0)
  11. Non-missing PreAdmRenalConsultCtr

Analysis Population

All patients from Population Two plus
  1. AKI stage 1 or greater (AKINStage >=1)
  2. No discharge diagnosis of ESRD within 30 days of admission (missing Discharge_ESRD_WI30)
  3. Not a preadmission hospice patient (PreAdmit_Hospice==0)
  4. First admission (FirstAKIAdmit = 1)



Variable Definitions

  • DaysFirst30DayPostPeakGTE60GFR - the first time after 30 days post peak in which the GFR for the patient was above or equal to 60.
    • take all of the creatinines for each patient after the first 30 days post peak
    • calculated their GFR based on the equation (race, age, gender)
    • filter to keep only those that were 60 or above
    • find the first time (in days after 30 days peak) in which that was the case
  • RecoveredAfter270Days - if patient's recovery (GFR of 60 or greater) is observed at or after 270 days in the after-30-days-post-peak period.
    • DaysFirst30DayPostPeakGTE60GFR >= 270
  • strataBlGFR60, two levels
    • "less than 60" if MostRecentOutPtGFR<60
    • "60 or more" if MostRecentOutPtGFR >= 60
  • HundrPercServConn
    • date of service connection is before admission: 100SCDate < admDate
  • Diagnosis Variables ("CADDx", "CHFDx", "DMDx", "HTNDx", "PVDDx")
    • CADDx TRUE (a person was diagnosed with CAD) if CADDxDate was equal or before the admission date (admDate)
  • Variable time-to-dialysis dialysisDays
    • minimum of PostPeak30DialysisCtrCD and PostPeak30DialysisCtrAppt



Analysis references

  1. Robert J. Gray, A CLASS OF K-SAMPLE TESTS FOR COMPARING THE CUMULATIVE INCIDENCE OF A COMPETING RISK, The Annals of Statistics, 1988, Vol 16, No. 3, 1141-1154
  2. Jason P. Fine, Robert J. Gray, A PROPORTIONAL HAZARD MODEL FOR THE SUBDISTRIBUTION OF A COMPETING RISK, Journal of the American Statistical Association, June 1999, Vol. 94, No. 446, 496-509
  3. Melania Pintilie, COMPETING RISKS, A PRACTICAL PERSPECTIVE, 2006, John Wiley & Sons, Ltd.
  4. Terry M. Therneau, Patricia M. Grambsch, MODELING SURVIVAL DATA, EXTENDING THE COX MODEL, 2000, Springer, New York


This topic: Main > Projects > ProjectsOfMichaelMatheny
Topic revision: revision 11
 
This site is powered by FoswikiCopyright © 2013-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback