
==============================================================================
PUBLISHED CREATININE GRID
==============================================================================
21 distinct values, top-coded at 5.54 mg/dL
[0.41, 0.47, 0.53, 0.58, 0.61, 0.65, 0.68, 0.71, 0.74, 0.77, 0.8, 0.84, 0.88, 0.92, 0.98, 1.05, 1.15, 1.33, 1.68, 3.03, 5.54]

adjacent spacing: min 0.03, median 0.04, max 2.51

reproduction check against gate1c: max |difference| = 0.000000 over 20 shared baselines

==============================================================================
BASELINE CREATININE DISTRIBUTION AND EFFECTIVE ABSOLUTE-RISE THRESHOLD
==============================================================================
 baseline  n_primary_cohort  pct_primary_cohort  kdigo_target  nearest_attainable  effective_rise_required  inflation_pct
     0.41                 9                0.38          0.71                0.71                     0.30            0.0
     0.47                34                1.44          0.77                0.77                     0.30            0.0
     0.53                48                2.03          0.83                0.84                     0.31            3.3
     0.58                57                2.41          0.88                0.88                     0.30            0.0
     0.61                57                2.41          0.91                0.92                     0.31            3.3
     0.65                91                3.85          0.95                0.98                     0.33           10.0
     0.68                96                4.06          0.98                0.98                     0.30           -0.0
     0.71               106                4.48          1.01                1.05                     0.34           13.3
     0.74               131                5.54          1.04                1.05                     0.31            3.3
     0.77               126                5.33          1.07                1.15                     0.38           26.7
     0.80               159                6.72          1.10                1.15                     0.35           16.7
     0.84               170                7.19          1.14                1.15                     0.31            3.3
     0.88               150                6.34          1.18                1.33                     0.45           50.0
     0.92               179                7.57          1.22                1.33                     0.41           36.7
     0.98               173                7.32          1.28                1.33                     0.35           16.7
     1.05               214                9.05          1.35                1.68                     0.63          110.0
     1.15               190                8.03          1.45                1.68                     0.53           76.7
     1.33               144                6.09          1.63                1.68                     0.35           16.7
     1.68               100                4.23          1.98                3.03                     1.35          350.0
     3.03                72                3.04          3.33                5.54                     2.51          736.7
     5.54                59                2.49          5.84                 NaN                      NaN            NaN

patients whose baseline is not a published grid value: 0

==============================================================================
PATIENT-WEIGHTED BURDEN OF THRESHOLD INFLATION
==============================================================================
                    inflation_band  grid_values  n_patients  pct_patients baseline_range
    at or near nominal (0.30-0.35)           10         799          33.8      0.41-0.84
   moderately inflated (0.35-0.50)            6         931          39.4      0.77-1.33
substantially inflated (0.50-1.00)            2         404          17.1      1.05-1.15
         severely inflated (>1.00)            2         172           7.3      1.68-3.03

                population  n_with_baseline_on_grid  pct_at_nominal_0.30  pct_inflated_above_0.30  pct_requiring_ge_0.45  pct_requiring_ge_0.60  median_effective_rise  q25_effective_rise  q75_effective_rise  mean_effective_rise
   primary cohort (n=2365)                     2306                  8.5                     91.5                   31.5                   16.7                   0.35                0.31                0.45                0.502
landmark risk set (n=1834)                     1775                  8.1                     91.9                   32.7                   17.6                   0.35                0.31                0.53                0.515

==============================================================================
IS THE MISCLASSIFICATION DIFFERENTIAL WITH RESPECT TO LACTATE?
==============================================================================

  Range of patient-weighted median effective rise across strata: 0.000 mg/dL
population  n_with_baseline_on_grid  pct_at_nominal_0.30  pct_inflated_above_0.30  pct_requiring_ge_0.45  pct_requiring_ge_0.60  median_effective_rise  q25_effective_rise  q75_effective_rise  mean_effective_rise
     <=1.9                      791                  7.2                     92.8                   33.2                   17.3                   0.35                0.33                0.53                0.512
   2.1-3.0                      557                  9.7                     90.3                   29.8                   15.8                   0.35                0.31                0.45                0.496
   3.8-5.9                      746                  9.5                     90.5                   30.2                   15.5                   0.35                0.31                0.45                0.493
8.6 capped                      212                  6.6                     93.4                   34.0                   21.2                   0.35                0.34                0.53                0.511

==============================================================================
PATIENTS FOR WHOM NEITHER KDIGO ARM IS ATTAINABLE
==============================================================================
baseline creatinine at the top code 5.54 mg/dL
  ratio arm requires >= 8.31, absolute arm >= 5.84; largest representable value is 5.54
  patients affected                    : 59 (2.5% of the primary cohort)
  observed AKI among them              : 0
  in the landmark risk set             : 59
  their observed peak creatinine (max) : 5.54

  distribution across lactate strata (n, and % of that stratum):
    <=1.9          24  (2.9%)
    2.1-3.0        13  (2.3%)
    3.8-5.9        19  (2.5%)
    8.6 capped      3  (1.4%)

==============================================================================
WHICH KDIGO ARM GENERATED THE EVENT, BY THRESHOLD-INFLATION BAND
==============================================================================
                    inflation_band  events  absolute_arm_involved  ratio_arm_involved  pct_absolute_arm  n_patients  aki_risk_pct
    at or near nominal (0.30-0.35)     219                  185.0               151.0              84.5         799          27.4
           criterion unsatisfiable       0                    0.0                 0.0               0.0          59           0.0
   moderately inflated (0.35-0.50)     312                  298.0               103.0              95.5         931          33.5
         severely inflated (>1.00)      73                   53.0                73.0              72.6         172          42.4
substantially inflated (0.50-1.00)     123                  113.0                63.0              91.9         404          30.4

==============================================================================
PRIMARY MODEL REFITTED WITHOUT THE STRUCTURALLY UNCOUNTABLE PATIENTS
==============================================================================
Those patients contribute person-time but cannot supply an event, so they
dilute the denominator. Refitting without them shows how much that matters.

                           analysis  n_at_risk  events  csHR[2.1-3.0] csHR[2.1-3.0]_ci  csHR[3.8-5.9] csHR[3.8-5.9]_ci  csHR[8.6 capped] csHR[8.6 capped]_ci  trend_HR  trend_ci  trend_p
as published (all at-risk patients)       1834     196          1.159        0.79-1.69          1.482        1.04-2.10             2.135           1.22-3.74     1.251 1.08-1.45   0.0034
 excluding structurally uncountable       1775     196          1.149        0.79-1.68          1.455        1.02-2.07             2.111           1.21-3.70     1.243 1.07-1.44   0.0046

written to results\manuscript_v3
