Donze et al developed a score for identifying a patient at risk for a potentially avoidable hospital readmission within 30 days. This can help to identify a patient who may benefit from more aggressive management. The authors are from Brigham and Women’s Hospital and Bern University Hospital.
Patient selection: medical patient who is hospitalized
Outcome: 30-day readmission
Parameters:
(1) H for hemoglobin at discharge
(2) O for oncology service
(3) S for serum sodium at dischage
(4) P is for procedure during admission (any ICD-9-CM coded)
(5) IT is for index admission type
(6) A is for number of hospital admissions during past year
(7) L is for length of stay in days
Parameter |
Finding |
Points |
hemoglobin at discharge |
>= 12 g/dL |
0 |
|
< 12 g/L |
1 |
oncology service |
no |
0 |
|
yes |
2 |
serum sodium |
>= 135 mmol/L |
0 |
|
< 135 mmol/L |
1 |
procedure |
none |
0 |
|
any |
1 |
index admission type |
elective |
0 |
|
nonelective |
1 |
number of hospital admissions |
0 |
0 |
|
1 to 5 |
2 |
|
6 or more |
5 |
length of stay |
< 5 days |
0 |
|
>= 5 days |
2 |
total score =
= SUM(points for all 7 parameters)
Interpretation:
• minimum score: 0
• maximum score: 13
• The higher the score the greater the risk of a potentially avoidable readmission.
Score |
Risk Category |
0 to 4 |
low (5%) |
5 or 6 |
intermediate (9-10%) |
7 to 13 |
high (18%) |
Performance:
• The c-statistic is 0.71.
Purpose: To identify a patient admitted to a medical service who is at risk for a potentially avoidable readmission using the HOSPITAL score of Donze et al.
Objective: criteria for diagnosis, selection
ICD-10: ,