|
| 1 | +--- |
| 2 | +title: MIMIC-II v2.6 |
| 3 | +parent: MIMIC-II documentation |
| 4 | +nav_order: 2 |
| 5 | +--- |
| 6 | + |
| 7 | +# MIMIC-II v2.6 |
| 8 | + |
| 9 | +## Changelog |
| 10 | + |
| 11 | +- Merge the last year of the ISM database |
| 12 | + |
| 13 | +The last year of ISM, through September 2008, has been added, including indications of matches to waveforms. |
| 14 | + |
| 15 | +- New dateshifts |
| 16 | + |
| 17 | +We have given all subjects a new, random dateshift. |
| 18 | + |
| 19 | +- Add overall SOFA score as total of each organ system's score |
| 20 | + |
| 21 | +The individual-system SOFA scores are in CHARTEVENTS, with itemids 20002 through 20007; a new itemid, 20009, has been added, and it s for the total score. The relevant columns in ICUSTAY\_DETAIL (sofa\_first, sofa\_min, sofa\_max) indicate this total score. |
| 22 | + |
| 23 | +- Add a column to ICUSTAY\_DETAIL indicating number of matched waveforms |
| 24 | + |
| 25 | +A new column, matched\_waveforms\_num, has been added to ICUSTAY\_DETAIL; it gives the number of waveforms matched to the patient. |
| 26 | + |
| 27 | +- Add waveform statistics |
| 28 | + |
| 29 | +Basic waveform statistics and metadata have been added to the database; these are essentially the organized output of what one would see by running the WFDB utility wfdbdesc (over all of the |
| 30 | +waveforms). |
| 31 | + |
| 32 | +- Add LOINC codes for hospital labs |
| 33 | + |
| 34 | +D\_LABITEMS contains two new columns, loinc\_code and loinc\_description, giving the LOINC codes and descriptions respectively. |
| 35 | + |
| 36 | +- Add diagnosis-related groups (DRGs) |
| 37 | + |
| 38 | +A new table, DRGEVENTS, maps subject IDs and hospital-admission IDs to DRG itemids, which are defined in D\_CODEDITEMS. |
| 39 | + |
| 40 | +- Add procedures |
| 41 | + |
| 42 | +PROCEDUREEVENTS contains indications of procedures performed on patients, for given hospital-admission IDs. The itemids for procedures are defined in D\_CODEDITEMS. Exact times for procedures |
| 43 | +are not given; however, their dates are given in the proc\_dt column, and the order in which a patient's procedures are performed, for a given day, are indicated by consecutive numbers (integers) in the sequence\_num column. |
| 44 | + |
| 45 | +- Add demographic information |
| 46 | + |
| 47 | +DEMOGRAPHICEVENTS contains indications of marital status, type and source of admission, ethnicity, overall payor group, and religion as recorded for each patient upon admission to the hospital. The |
| 48 | +definitions of the itemids are given in D\_CODEDITEMS. |
| 49 | + |
| 50 | +- Add microbiology information |
| 51 | + |
| 52 | +MICROBIOLOGYEVENTS contains information on microbiology tests performed on patients: specimen, organism, and antibacterium are indicate in the spec\_itemid, org\_itemid, and ab\_itemid columns |
| 53 | +respectively; these itemids are defined in D\_CODEDITEMS. |
| 54 | + |
| 55 | +- Remove duplicate mappings of waveform case IDs to subject IDs in PARAMETER\_MAPPING |
| 56 | + |
| 57 | +- Fix the icustay\_expire\_flg in ICUSTAY\_DETAIL |
| 58 | + |
| 59 | +icustay\_expire\_flg indicates whether the given patient died during the given ICU stay. For some patients, this column was 'Y' for all of his/her ICU stays, rather than just the last; this has been fixed. |
| 60 | + |
| 61 | +- Fix case of hospital admission ID not unique to subject ID |
| 62 | + |
| 63 | +Hospital admission 15449 was mapped to two different patients; this has been fixed. |
| 64 | + |
| 65 | +- Add Exlihauser comorbidity scores |
| 66 | + |
| 67 | +The Exlihauser comorbidity scores (calculated from DRG and ICD-9 codes) can be found in COMORBIDITY\_SCORES |
| 68 | + |
| 69 | +## Known limitations, bugs. |
| 70 | + |
| 71 | +- Missing data |
| 72 | + |
| 73 | +Newly added data: DRGs, Demographics, Microbiology tests and procedures were not available for all subjects. Therefore, you will find that this data is missing for some subjects. |
| 74 | + |
| 75 | +We were unable to complete the import of some of the text |
| 76 | +data. Radiology, Echo and ECG reports are missing from NOTEEVENTS. |
| 77 | + |
| 78 | +- Waveform dates |
| 79 | + |
| 80 | +The waveform metadata contains dates from the current waveforms available on PhysioNet: |
| 81 | + |
| 82 | +https://archive.physionet.org/physiobank/database/mimic2db/ |
| 83 | + |
| 84 | +The dates found here will not match those found elsewhere in the |
| 85 | +database. |
| 86 | + |
| 87 | +# MIMIC-II v2.5 |
| 88 | + |
| 89 | +## Changelog |
| 90 | + |
| 91 | +Major changes since version 2.4 : |
| 92 | + |
| 93 | +- Date of death after hospital discharge |
| 94 | + |
| 95 | +Patients have been matched against the Social Security death tapes, and the date of death has been imported into MIMIC-II. For patients who died in the hospital, their respective dates of discharge were used as their dates of death. Two additional columns have been added to the D\_PATIENTS table: |
| 96 | + |
| 97 | +- DOD = Date of Death |
| 98 | +- HOSPITAL\_EXPIRE\_FLG = Whether or not the patient deid in the hospital |
| 99 | + |
| 100 | +The column EXPIRE\_FLG in the ADMISSION table has been removed. |
| 101 | + |
| 102 | +- New Waveform case ID matches |
| 103 | + |
| 104 | +A first round of waveforms with up to 8 simultaneous channels have been matched to SUBJECT\_IDs. The matches can be found in the PARAMETER\_MAPPING table where CATEGORY is "SUBJECT\_ID\_TO\_CASE\_ID", the columns PARAM1\_NUM (for the SUBJECT\_ID) and PARAM2\_STR (for the waveform case ID). |
| 105 | + |
| 106 | +- Merge NOTEEVENTS into REPORTEVENTS |
| 107 | + |
| 108 | +REPORTEVENTS has been merged into NOTEEVENTS. Categories have been extracted into D\_REPORTITEMS, with the ITEMID column as the foreign key, but not completely; the D\_* tables will be combined into a complete dictionary in a later release. |
| 109 | + |
| 110 | +- New ICUSTAY\_DAYS materialized view |
| 111 | + |
| 112 | +The ICUSTAY\_DAYS materialized view was added to divide each ICU stays into daily periods. This greatly helps computation of cumulative queries. |
| 113 | + |
| 114 | +- Daily SOFA scores |
| 115 | + |
| 116 | +New entries in D\_CHARTITEMS were added to represent SOFA scores (itemid = 20002), and corresponding scores were added in CHARTEVENTS. |
| 117 | + |
| 118 | +- Daily SAPS I calculations |
| 119 | + |
| 120 | +For each ICU stay, a daily SAPS I was added into the CHARTEVENTS table, with ITEMID 20001. |
| 121 | + |
| 122 | +- Add a view for summarizing the D\_CHARTITEMS table |
| 123 | + |
| 124 | +It was somewhat troublesome for users of the database to collect statistics for the parameters (ITEMIDs) in D\_CHARTITEMS. Therefore, a summarizing materialized view, D\_CHARTITEMS\_DETAIL, was added as a first introduction of all charted parameters; it describes how many patients each parameter was charted from, and various statistics such as min., max., and average. |
| 125 | + |
| 126 | +- Clean "value1" and "value2" columns in CHARTEVENTS for numeric itemids |
| 127 | + |
| 128 | +Some ITEMIDs in the CHARTEVENTS table that are supposed to be only numeric (such as ABP, HR) had repeated values in columns VALUE1/VALUE2 (which are VARCHAR columns) and VALUE1NUM/VALUE2NUM (NUMBER columns); VALUE1 and VALUE2 were cleaned accordingly. |
| 129 | + |
| 130 | +- Reorganized ICUSTAY\_DETAIL materialized view |
| 131 | + |
| 132 | +The ICUSTAY\_DETAIL view has been used as a key component in many of computation for several researches. We added/changed some columns to reflect: |
| 133 | + |
| 134 | +- EXPIRE\_FLG = Flag to indicate if the paitent is dead, either in the hospital or after discharge |
| 135 | +- HOSPITAL\_EXPIRE\_FLG = If the patient deid in the hospital |
| 136 | +- ICUSTAY\_EXPIRE\_FLG = If the patient deid in the ICU |
| 137 | +- HEIGHT = Patient height |
| 138 | +- WEIGHT\_FIRST = The first measured weight for the patient in the ICU stay |
| 139 | +- WEIGHT\_MIN = The minimum weight for the ICU stay |
| 140 | +- WEIGHT\_MAX = The maximum weight for the ICU stay |
| 141 | +- SAPSI\_FIRST = The first measured SAPS-I score for the patient in the ICU stay |
| 142 | +- SAPSI\_MIN = The minimum SAPS-I score for the ICU stay |
| 143 | +- SAPSI\_MAX = The maximum SAPS-I score for the ICU stay |
| 144 | +- SOFA\_FIRST = The first measured SOFA score for the patient in the ICU stay |
| 145 | +- SOFA\_MIN = The minimum SOFA score for the ICU stay |
| 146 | +- SOFA\_MAX = The maximum SOFA score for the ICU stay |
| 147 | + |
| 148 | +- Add an ICUSTAY\_ID column to all event tables |
| 149 | + |
| 150 | +To facilitate queries and other computations, the ICUSTAY\_ID column has been added to event tables: CHARTEVENTS, IOEVENTS, MEDEVENTS, LABEVENTS and TOTALBALEVENTS. |
| 151 | + |
| 152 | +- Clean duplicates in D\_CAREUNITS |
| 153 | + |
| 154 | +There were duplicate names in the table D\_CAREUNITS that needed to be merged, and their foreign keys updated. |
| 155 | + |
| 156 | +- Remove entries from item tables which contain no records in event tables |
| 157 | + |
| 158 | +292 unused rows have been removed from D\_MEDITEMS; 3544 from D\_CHARTITEMS. |
| 159 | + |
| 160 | +- Fix problems with duplicate hospital admissions |
| 161 | + |
| 162 | +Six patients had hospital admissions with overlapping admission-to-discharge date intervals. These were fixed by examining the discharge summaries. |
| 163 | + |
| 164 | +- Add missing foreign key for POE\_ORDER |
| 165 | + |
| 166 | +For POE\_ORDER, columns SUBJECT\_ID and HADM\_ID were properly made to be foreign keys for the tables D\_PATIENTS and ADMISSIONS, respectively. |
| 167 | + |
| 168 | +## Database statistics (adult records summary) |
| 169 | + |
| 170 | +| | N* | #cases** (%unit) | mean ±std | median | IQR (Q1,Q3) | |
| 171 | +|---- | ---- | ---- | ---- | ---- | ---- | |
| 172 | +|Hospital admissions | 22,870 | | | | | |
| 173 | +|ICU Stays | 25,328 | | | | | |
| 174 | +|Age (years) | 25,328 | | 63.3 ±17.7 | 65.20 | 26.1 (51.3,77.4)| |
| 175 | +|Gender (male) | 25,328 | 14,259 (56.3%) | | | | |
| 176 | +|ICU Los (days) | 25,328 | | 4.7 ±11 | 2.20 | 3.3 (1.1,4.4)| |
| 177 | +|Hospital Los (days) | 22,870 | | 10.7 ±13 | 7.00 | 9 (4,13)| |
| 178 | +|SAPS I - day 1 *** | 20,142 | | 13.4 ±5.3 | 13.00 | 7 (10,17)| |
| 179 | +|Mechanical ventilation | 25,328 | 11,969 (47.3%) | | | | |
| 180 | +|Invasive Swan-Ganz hemodynamic monitoring | 25,328 | 5,637 (22.3%) | | | | |
| 181 | +|Invasive arterial blood pressure monitoring | 25,328 | 14,054 (55.5%) | | | | |
| 182 | +|Use of vaso-active medications | 25,328 | 8,693 (34.3%) | | | | |
| 183 | +|Hospital mortality | 22,870 | 2,666 (11.7%) | | | | |
| 184 | + |
| 185 | +\* N = The number of ICUStays or Hospitalizations. that were included in the calculation for the given ICU Unit. |
| 186 | + |
| 187 | +** #cases = The number of ICUStays orHospitalizations that match the category criteria. For example: In Gender, #cases will be the number of male ICUStays in a unit. |
| 188 | + |
| 189 | +*** Only ICUStays where they have all 14 SAPS parameters were included |
| 190 | + |
| 191 | +# MIMIC-II v2.4 |
| 192 | + |
| 193 | +## Changelog |
| 194 | + |
| 195 | +- Bug #5: Fix problems with icustay generation |
| 196 | +- Bug #13: Fix in "Unknown" service type in icustay\_detail |
| 197 | +- Bug #21: Rename ORDER column in PARAMETER\_MAPPING table |
| 198 | +- Bug #26: Make event date fields consistent |
| 199 | +- Bug #27: Rename d\_labitems.lab\_category to category for consistency with other item tables |
| 200 | +- Bug #40: Modify SAPS I calculation to correctly account for ventilated patients |
| 201 | +- Feature #11: Delete the materialized view VITAL\_SIGNS\_RAW, because this was just a temporal data to be used by some statistics calculations. |
| 202 | +- Feature #12: Add indices on cgid, itemid to chartevents |
| 203 | +- Feature #14: Add AKI materialized view |
| 204 | +- Feature #15: Add organ system failure materialized view |
| 205 | +- Feature #16: Restore the Physionet 2009 challenge patients |
| 206 | +- Feature #34: Re-generate SAPS scores after fixing Icustay problems |
| 207 | +- Change request #36: Make the "Length Of Stay" columns consistent in minutes |
| 208 | +- Change request #37: Replace ICUSTAY materialized view with ICUSTAYEVENTS table |
| 209 | +- Change request #38: Make consistent some column names in ICUSTAY_DETAIL |
| 210 | + |
| 211 | +## Database statistics (adult records summary) |
| 212 | + |
| 213 | +| | N* | #cases** (%unit) | mean ±std | median | IQR (Q1,Q3) | |
| 214 | +|---- | ---- | ---- | ---- | ---- | ----| |
| 215 | +|Hospital admissions | 22,870 | | | | | |
| 216 | +|ICU Stays | 25,328 | | | | | |
| 217 | +|Age (years) | 25,328 | | 63.3 ±17.7 | 65.20 | 26.1 (51.3,77.4)| |
| 218 | +|Gender (male) | 25,328 | 14,259 (56.3%) | | | | |
| 219 | +|ICU Los (days) | 25,328 | | 4.7 ±11 | 2.20 | 3.3 (1.1,4.4)| |
| 220 | +|Hospital Los (days) | 22,870 | | 10.7 ±13 | 7.00 | 9 (4,13)| |
| 221 | +|SAPS I - day 1 *** | 20,142 | | 13.4 ±5.3 | 13.00 | 7 (10,17)| |
| 222 | +|SAPS I - day 2 *** | 13,495 | | 11.8 ±4.6 | 11.00 | 6 (9,15)| |
| 223 | +|SAPS I - day 3 *** | 9,070 | | 12.2 ±4.5 | 12.00 | 6 (9,15)| |
| 224 | +|Mechanical ventilation | 25,328 | 11,969 (47.3%) | | | | |
| 225 | +|Invasive Swan-Ganz hemodynamic monitoring | 25,328 | 5,637 (22.3%) | | || |
| 226 | +|Invasive arterial blood pressure monitoring | 25,328 | 14,054 (55.5%) | | || |
| 227 | +|Use of vaso-active medications | 25,328 | 8,693 (34.3%) | | || |
| 228 | +|Hospital mortality | 22,870 | 2,666 (11.7%) | | || |
| 229 | + |
| 230 | +\* N = The number of ICUStays or Hospitalizations. that were included in the calculation for the given ICU Unit. |
| 231 | + |
| 232 | +** #cases = The number of ICUStays orHospitalizations that match the category criteria. For example: In Gender, #cases will be the number of male ICUStays in a unit. |
| 233 | + |
| 234 | +*** Only ICUStays where they have all 14 SAPS parameters were included. 2nd and 3rd day data comes from version 2.5 of the database. |
0 commit comments