time interval used in kdigo-uo.sql

To calculate urine ouput over 3 periods, that is:

  -- three sums:
  -- 1) over a 6 hour period
  -- 2) over a 12 hour period
  -- 3) over a 24 hour period

The code in mimic-code/concepts/organfailure/kdigo-uo.sql is:

  , sum(case when iosum.charttime <= io.charttime + interval '5' hour
      then iosum.VALUE
    else null end) as UrineOutput_6hr
  , sum(case when iosum.charttime <= io.charttime + interval '11' hour
      then iosum.VALUE
    else null end) as UrineOutput_12hr
  , sum(iosum.VALUE) as UrineOutput_24hr
....
and iosum.charttime <= (io.charttime + interval '23' hour)
....

The 3 intervals used were 5/11/23 hrs. And later in kdigo-stages-48hr.sql and kdigo-stages-7days.sql:

..
min(uo.urineoutput_6hr / uo.weight / 6.0)::numeric as uo6
...
, min(uo.urineoutput_12hr / uo.weight / 12.0)::numeric as uo12
...
, min(uo.urineoutput_24hr / uo.weight / 24.0)::numeric as uo24
...

so, shouldn't the above intervals be 6/12/24 hrs?

And in the comments from kdigo-stages-48hr.sql and kdigo-stages-7days.sql:

-- For urine output: the highest UO in hours 0-48 is used

Aren't we using the lowest UO as seen from the code above min()?

Correct me if I am missing something here, thanks.

Start with mimic-code/concepts/organfailure/kdigo-uo.sql, then compare the interval calculations with kdigo-stages-48hr.sql and kdigo-stages-7days.sql. Check how the documented 6-, 12-, and 24-hour windows relate to the 5/11/23-hour bounds and whether min() matches the comments about the highest urine output. Done means the SQL behavior and its comments consistently express the intended KDIGO calculation.