APACHE physiological variable collection

When to collect, which value to use

See Selection and timing of APACHE components

Exceptionally high or low values

CCMDB.accdb has error checks for values that are unlikely high or low to prevent errors. The ranges were determined by the statistician based on previous data(see abs min and abs max in APACHE_Scoring_table). To keep the checks meaningful the threshold values will not be changed for extreme outliers.

For values outside the "warning" limit, you must enter the reason for this into the Notes field so that we can validate that these were not errors.

For values outside the "absolut" limit, enter the closest value allowed and enter a comment into the Notes field. Also, email Pagasa the correct value.

Guidelines for the Values to use for APACHE II Physiologic Variables

Use the worst

• The general rule is to use the worst value as per APACHE II rules in the initial 24 hours of the ICU record (remember that ICU records begin when the patient begins being cared for by the ICU service, i.e. first Service tmp entry DtTm.

When not available

The following info is relevant if no measurements were taken or in cases of Lost/missing chart.

• You can not enter 0 for any of these, and you can't leave them blank.
• If for a given parameter there are no values during that initial 24 hours after the start of the profile, you can go back prior to the start of the ICU record, and use the value closest to the start time of the ICU record (but max 24 hrs prior to admission).
  • If there are no values for that parameter either during the first 24 hours OR up to 24 hours prior to the start of the ICU record, APACHE assumes the value was normal -- so code the value that is the midpoint of the normal range per APACHE rules.

Values during Arrest

• DO NOT use vital signs from arrests. Instead use the worst values from recordings OTHER THAN the code sheet.
• The rationale is that codes are specific, usually short interludes and if it occurs there WILL be an ICD10 diagnosis listed for it (Cardiac arrest).
• You can use labs drawn during arrests, but only do so if these are the ONLY labs available.
• For the lab data, you can use values both from the central laboratory and those done on a bloodgas machine.

Iatrogenically changed values

• Keep in mind the diagnosis and try not to include iatrogenically modified values, e.g. BP severely high due to inotropes or decreased LOC due to sedation.
• However, if a value is assessed as part of testing whether an intervention works, that value is not necessarily ineligible (e.g. see Arterial_blood_gas_(APACHE)#ABGs_as_part_of_apnea_tests).