Readmissions to Intensive Care: a Prospective Multicenter Study in Australia and New Zealand

Supplemental Digital Content

Readmissions to Intensive Care: A prospective multicenter study in Australia and New Zealand

Authors

John D Santamaria MD,

Graeme J Duke MD

David V Pilcher MD

D James Cooper MD

John Moran MD

Rinaldo Bellomo MD

Participating Hospitals

New Zealand:
Auckland City Hospital (Cardio Thoracic & Vascular ICU), Auckland; Auckland City Hospital, Department Critical Care Medicine, Auckland; Christchurch Hospital, Christchurch; Hawke’s Bay Hospital, Hastings; Middlemore Hospital, Papatoetoe; Tauranga Hospital, Tauranga; Wellington Hospital, Wellington.

Australia:
ACT: Canberra Hospital, Garran.
NSW: Blacktown Hospital, Blacktown; ,Concord Repatriation General Hospital, Concord; John Hunter Hospital, New Lambton; Lismore Base Hospital, Lismore; Liverpool Hospital, Liverpool; Nepean Hospital, Kingswood; Prince of Wales Hospital, Randwick; Royal Prince Alfred Hospital, Camperdown; St George Hospital, Kogarah; St Vincent’s Hospital Darlinghurst; Westmead Hospital, Westmead.
NT: Royal Darwin Hospital, Darwin.
QLD: Townsville Hospital, Douglas; Wesley Hospital, Auchenflower.
SA: Flinders Medical Centre, Bedford Park; Queen Elizabeth Hospital, Woodville South;
TAS: Royal Hobart Hospital, Hobart.
VIC: Alfred Hospital, Melbourne; Austin Hospital, Heidelberg; Bendigo Hospital, Bendigo; Box Hill Hospital, Box Hill; Central Gippsland Hospital, Sale; Frankston Hospital, Frankston; Geelong Hospital, Geelong; Monash Medical Centre, Clayton; Royal Melbourne Hospital, Melbourne; St Vincent’s Hospital Melbourne, Fitzroy; The Northern Hospital, Epping; Western Hospital, Footscray.
WA: Fremantle Hospital, Fremantle; Royal Perth Hospital, Perth; Sir Charles Gairdner Hospital, Nedlands.

Legend of Supplemental Digital Content

Table E1. Characteristics of the participating Intensive Care Units

Table E2: Data Dictionary for Discharge & Readmission Evaluation (DARE) Study

Table E3. Primary diagnosis category on admission to ICU (APACHE III, major groups)

Table E4. Twenty (20) most common primary admission diagnoses (APACHE III)

Table E5. Readmission rates according to primary admission diagnosis

Figure E1. Distribution of days between original ICU discharge and readmission to ICU according to planned or unplanned status.

Figure E2. Hour of readmission to ICU according to planned or unplanned readmission.

Figure E3. Interval in days to first readmission stratified by subsequent survival or mortality.

Table E6. Comparison of early vs late readmission

Table E7. Comparisons between patients not readmitted (9626) and first readmission to ICU (581).

Table E8. Diagnoses on first and second ICU admission in 581 readmitted patients

Table E9 Results of fixed effects logistic regression predicting unplanned readmission following the first ICU discharge

Table E10. Estimates for the frailty model using STRMCURE.

References

Table E1 Characteristics of the participating Intensive Care Units

New Zealand

Hospital beds / ICU beds / ICU admits / Readmission
Rate / Mortality
Rate / Cases / MET / Liais / Trach
NZ01 / >500 / 14 / 1800 / 5.3% / 4.3% / 300 / Y / Y / Y
NZ02 / >500 / 14 / 1100 / 5.9% / 1.1% / 463 / N / N / N
NZ03 / >500 / 12 / 1100 / 5.7% / 3.0% / 297 / N / Y / N
NZ04 / 200-500 / 11 / 1000 / 4.3% / 3.1% / 161 / N / N / N
NZ05 / >500 / 12 / 1000 / 2.7% / 3.5% / 258 / Y / Y / N
NZ06 / 200-500 / 5 / 500 / 0% / 9.4% / 85 / N / N / Y
NZ07 / 200-500 / 14 / 1300 / 5.7% / 3.1% / 262 / N / Y / N

Australia

Hospital beds / ICU beds / ICU admits / Readmission / Mortality / Cases / MET / Liais / Trach
AU01 / >500 / 15 / 1500 / 4.7% / 3.1% / 404 / Y / Y / Y
AU02 / 200-500 / 16 / 1000 / 8.2% / 8.2% / 158 / Y / N / N
AU03 / 200-500 / 10 / 900 / 6.9% / 6.5% / 216 / Y / N / Y
AU04 / >500 / 14 / 1200 / 11.2% / 7.4% / 188 / Y / Y / Y
AU05 / 200-500 / 9 / 700 / 0% / 7.5% / 107 / Y / Y / N
AU06 / >500 / 26 / 2000 / 4.3% / 8.1% / 185 / Y / Y / Y
AU07 / 200-500 / 17 / 1300 / 5.3% / 5.8% / 225 / Y / Y / N
AU08 / >500 / 48 / 3400 / 0.8% / 0.0% / 123 / Y / Y / Y
AU09 / >500 / 10 / 1300 / 6.2% / 7.6% / 145 / N / N / N
AU10 / >500 / 15 / 900 / 6.6% / 9.9% / 182 / N / Y / Y
AU11 / 200-500 / 15 / 1300 / 8.7% / 4.6% / 173 / N / N / N
AU12 / >500 / 20 / 1000 / 2.4% / 11.1% / 208 / Y / N / N
AU13 / 200-500 / 14 / 900 / 5.8% / 0.7% / 139 / N / Y / N
AU14 / 200-500 / 13 / 1000 / 1.7% / 2.5% / 359 / Y / N / N
AU15 / 200-500 / 19 / 1600 / 6.2% / 2.9% / 275
AU16 / >500 / 29 / 2100 / 7.0% / 7.2% / 471 / Y / N / N
AU17 / 200-500 / 14 / 900 / 7.3% / 8.9% / 191 / Y / Y / Y
AU18 / >500 / 17 / 1200 / 4.7% / 3.8% / 442 / Y / Y / Y
AU19 / >500 / 18 / 2100 / 7.7% / 7.0% / 442 / Y / Y / Y
AU20 / 200-500 / 28 / 1800 / 9.2% / 4.7% / 510 / Y / Y / Y
AU21 / 200-500 / 7 / 600 / 5.6% / 4.0% / 124 / Y / Y / N
AU22 / 200-500 / 9 / 700 / 6.8% / 10.5% / 190 / N / Y / N
AU23 / 200-500 / 6 / 300 / 4.8% / 4.8% / 83 / Y / Y / N
AU24 / 200-500 / 8 / 800 / 3.4% / 1.7% / 174 / Y / Y / N
AU25 / 200-500 / 11 / 1400 / 5.4% / 3.0% / 331 / Y / Y / Y
AU26 / >500 / 19 / 1600 / 4.2% / 6.4% / 312 / Y / Y / N
AU27 / 200-500 / 10 / 800 / 7.0% / 6.0% / 199 / Y / Y / Y
AU28 / 200-500 / 24 / 2100 / 7.1% / 6.7% / 449 / Y / Y / N
AU29 / 200-500 / 11 / 1100 / 5.7% / 5.0% / 565 / Y / Y / Y
AU30 / 200-500 / 12 / 800 / 6.0% / 7.3% / 150 / Y / Y / N
AU31 / >500 / 18 / 1400 / 7.4% / 6.7% / 299 / Y / N / N
AU32 / 200-500 / 12 / 800 / 0% / 5.7% / 105 / N / Y / N
AU33 / >500 / 31 / 2700 / 4.6% / 6.2% / 389 / Y / N / N

Cases: number of cases provided in the study. MET: presence of medical emergency or rapid response team in the hospital run by ICU. Lias: presence of nurses from ICU specifically to follow patients on the ward. Trach: multidisciplinary team to follow patients with tracheostomy on ward.

There were 5 units situated in regional centres (where short term mechanical ventilation and even renal replacement therapy would be considered); 9 in metropolitan centres (where medium to long term ventilation, renal replacement therapy would be undertaken, and transfers made for specialist cardiac surgery, neurosurgery etc.); 17 tertiary units (ventilation, renal replacement, cardiac surgery and neurosurgery with occasional transfers for quaternary services); 9 centres offering specific quaternary services that might include heart/lung transplant, liver transplant, burns, bone marrow transplantation. Thirty-five of ICUs had University affiliation.

Table E2: Data Dictionary for Discharge & Readmission Evaluation (DARE) Study

1 / ICU Discharge Number / Number / Unique number every new discharge from ICU for the study
2 / Patient Identifier / Number/Text / Unique De-identified number linked to MRN on Unit worksheet. Numbers of text appropriate but unique for that patient.
3 / ICU Admission Date/Time / Date/Time / Date/Time of admission to ICU.
DDMMYY, HHMM
4 / ICU Admission Origin / Tick one box / ED if from Emergency, OR if from Operating Theatre or Recovery Room, Ward if from other hospital ward, or Tx if transferred directly from another institution/hospital
5 / Patient Deemed Ready to Leave ICU. Date/Time / Date/Time / Date/Time when attending staff determine that the patient is ready for transfer
DDMMYY, HHMM
6 / Actual ICU Discharge Date/Time / Date/Time / Actual Date/Time when patient was discharged from ICU
DDMMYY, HHMM
7 / Was Discharge from ICU Premature / Tick if Yes / In the opinion of the intensivist/team, would the patient have benefitted by remaining in ICU longer?
8 / Invasive Ventilation / Tick if Yes / Mechanical ventilation via ETT or tracheostomy
9 / Non-invasive Ventilation / Tick if Yes / Ventilation via nasal or full face mask. Excludes CPAP for sleep apnoea, high flow nasal oxygen
10 / Inotrope Infusion / Tick if Yes / Infusions of inotropes for augmentation of blood pressure or cardiac output
11 / Renal replacement therapy / Tick if Yes / Hemofiltration, Hemodialysis etc for renal impairment
Excludes therapeutic plasmafiltration.
12 / IABP / Tick if Yes / Intra-aortic balloon pump
13 / Other Invasive Therapy / Tick if Yes and comment / Other advanced therapies e.g. ECMO, oscillation, MARS
14 / Not for Resuscitation Order / Tick if Yes / Written documentation in history that patient is not for Resuscitation, CPR, etc
15 / Not for Return to ICU / Tick if Yes / Written or verbal order that patient will not be readmitted to ICU in the event of deterioration
16 / Not for cardiac arrest team / Tick if Yes / Written order that patient should not have the cardiac arrest team called in the event of deterioration
17 / Not for emergency (MET) Call / Tick if Yes / Written order that patient should not have MET call in the event of deterioration where a MET service exists
18 / Tracheostomy / Tick if Yes / Tracheostomy in place on discharge from ICU
19 / Non-invasive ventilation / Tick if Yes / Ventilation via nasal or full face mask on discharge from ICU. Excludes CPAP for sleep apnoea, high flow nasal oxygen
20 / Inotrope Infusion / Tick if Yes / Infusions of inotropes for augmentation of blood pressure or cardiac output on discharge from ICU
21 / Ongoing Renal Replacement / Tick if Yes / On discharge from ICU, it is planned that the patient will continue with dialysis (HD, PD, HF etc)
22 / TPN / Tick if Yes / On discharge, patient continues or will receive TPN
23 / Altered conscious state / Tick if Yes / On discharge, any departure from normal conscious state (GCS=15). Includes confusional state
24 / Glasgow Coma Score on discharge from ICU / Number / Total GCS around time of discharge. If patient has tracheostomy, estimate if they can talk normally for full verbal points
25 / Written Medical Discharge Summary / Tick if Yes / On discharge, a written medical discharge summary is in the medical record
26 / Written Nursing Discharge Summary / Tick if Yes / On discharge, a written nursing discharge summary is in the medical record
27 / Medical Handover Documented / Tick if Yes / On discharge, written documentation of handover to medical staff on ward
28 / Nursing Handover Documented / Tick if Yes / On discharge, written documentation of handover to nursing staff on ward
29 / Date/Time of first observations on ward / Date/Time / Date and time (actual or approximate) of the first documented observations (e.g. temp, pulse, BP) on ward after discharge from ICU
30 / Date/Time of first medical notes / Date/Time / Date and time (actual or approximate) of first notes by primary/parent medical/surgical team after discharge from ICU.
31 / Date/Time of first visit by senior consultant / Date/Time / Date and time (actual or approximate) of first documented visit by senior consultant (physician, surgeon, psychiatrist etc) after discharge from ICU.
32 / Patient was discharged from ICU to / Tick if Yes / Tick only one box
Discharged to high dependency unit or stepdown when nurse:patient is 1:2
Discharged to specialty ward such as neurosurgery where nurse:patient ratios are the same as general areas
Discharged to general medical/surgical/obstetric ward where nurse:patient ratios are standard
If the patient is a readmission and dies, goes home or to another hospital, leave this section blank.
33 / Number of emergency (MET) calls / Number / Number of emergency calls between first discharge from ICU and Hospital discharge. Does NOT include cardiac arrest calls
34 / Number of cardiac arrest calls / Number / Number of cardiac arrest calls between first discharge from ICU and Hospital discharge. Patient required CPR ± defibrillation
35 / Number of Liaison Visits / Number / Number of formal nurse/medical liaison visits between first discharge from ICU and Hospital discharge
36 / Age / Number / Age of Patient in years
37 / Gender / Text / M|F
38 / Hospital Admission Date and Time / Date/Time / Date /Time of admission to Hospital.
Can be obtained from AORTIC or equivalent
39 / Hospital Admission Source / Tick / Tick one only of Home, Other Acute Hospital, Other Chronic Hospital (including residential aged care, nursing home), or Other ICU
40 / APACHE III risk of death / Number / Calculated APACHE Risk of Death from AORTIC or equivalent for first 24 hours of admission.
If this is not collected electronically, complete the APACHE III calculation sheet. Risk of death will be calculated from the scores and the Diagnostic Code (below)
41 / ANZICS APACHE III Diagnostic Code / Number / The ANZICS modification to the APACHE III diagnostic codes. See APACHE III Worksheet.
42 / Hospital Discharge Date and Time / Date/Time / Date/Time of discharge from Hospital.
Includes discharge to rehabilitation, mental health, palliative care.
Can be obtained from AORTIC or equivalent
43 / Hospital Outcome / Tick / Tick one only of Still in Hospital (at study termination date), Died in Hospital, Discharged home, Transfer to Rehabilitation, Transfer to other ICU, or Transfer to other Hospital
44 / Proximate Cause of Death / Tick / Tick one only for the immediate reason that the patient died.
45 / Underlying Causes of Death / Tick / Tick the boxes for
46 / If the patient left ICU without an NFR order AND subsequently died, was an NFR form completed later in the ward? / Tick / Tick YES if, following discharge from ICU, the patient subsequently had a written not for resuscitation form completed and placed in the medical record.

Readmission Questions

Table E3. Primary diagnosis categories on admission to ICU (APACHE III, major groups)