Manzano et al reported the Cancer READMIT score for predicting the risk of unplanned 30-day readmission for patient with cancer. This can help to identify a patient who may benefit from more aggressive management after discharge. The authors are from The University of Texas MD Anderson Cancer Center.
Cancer READMIT is derived from cancer, recent ER, English, anemia, days LOS, marital status, increased WBC, tumor spread.
Parameter:
(1) readmission risk based on cancer site
(2) ER visit in the 30 days before index hospitalization
(3) primary language
(4) discharge hemoglobin for index hospitalization in g/dL
(5) length of stay in days
(6) marital status
(7) discharge WBC for index hospitalization, in 10^9/L
(8) distant metastases
|
Parameter
|
Finding
|
Points
|
|
cancer site
|
low risk
|
0
|
|
|
high risk
|
3
|
|
ER visit
|
no
|
0
|
|
|
yes
|
2
|
|
primary language
|
English
|
0
|
|
|
other
|
2
|
|
discharge hemoglobin
|
< 10 g/dL
|
1
|
|
|
>= 10 g/dL
|
0
|
|
length of stay
|
<= 4 days
|
0
|
|
|
> 4 days
|
2
|
|
marital status
|
married or with partner
|
0
|
|
|
not married
|
1
|
|
discharge WBC
|
<= 11 * 10^9/L
|
0
|
|
|
> 11 * 10^9/L
|
1
|
|
distant metastases
|
no
|
0
|
|
|
yes
|
2
|
where:
• Cancer sites that are low risk for readmission: breast, skin, genitourinary or endocrine
• Cancer sites that are high risk: bone, soft tissue, CNS, GI, gynecologic, head and neck, lung, other thoracic, unknown primary.
total score =
= SUM(points for all of the parameters)
Interpretation:
• minimum score: 0
• maximum score: 14
• The higher the score the greater the risk of 30-day unplanned readmission.
|
Score
|
Risk Readmission
|
|
0 to 4
|
low
|
|
5 to 7
|
medium
|
|
>= 8
|
high
|
Performance:
• The area under the ROC curve in the derivation cohort was 0.65 (bias-corrected) and in the validation set was 0.63.
