Han et al reported a nomogram for predicting readmission for a patient with chronic heart failure. This can help to identify a patient who may require more aggressive management. The authors are from Shanxi Medical University in China.
Analysis: CHF-PROM questionnaire, with 12 subdomains and 57 questions, scored from 0 (never) to 4 (always)
Parameters:
(1) gender
(2) healthcare
(3) income
(4) appetite and sleep from CHF-PROM, from 2 to 16 (4 questions)
(5) independence from CHF-PROM, from 0 to 16 (4 questions)
(6) anxiety from CHF-PROM, from 0 to 35 (8 questions)
(7) depression from CHF-PROM, from 0 to 24 (6 questions)
(8) paranoia, from CHF-PROM, from 4 to 16 (4 questions)
(9) support from CHF-PROM, from 0 to 20 (5 questions)
points for anxiety =
= (2.857 * (anxiety level))
points for appetite and sleep =
= (3.19 * (points for appetite)) - 6.38
points for depression =
= (1.54 * (depression level))
points for independence =
= (4.64 * (independence level))
points for paranoia =
= 129.6 - (8.1 * (paranoia level))
points for support =
= (3.255 * (support level))
Parameter
|
Finding
|
Points
|
gender
|
female
|
0
|
|
male
|
24
|
healthcare
|
urban
|
0
|
|
rural
|
21.3
|
|
self-pay
|
82.8
|
income
|
high
|
0
|
|
middle
|
44.5
|
|
low
|
65.6
|
total score =
= SUM(points for all of the parameters)
Interpretation:
• minimum score: 0
• maximum score: 590.5
value of X =
= (0.01634 * (score)) - 3.89
probability of being without readmission at 1 year =
= 1 / (1 + EXP((-1) * X))
probability of readmission at 1 year =
= 1 - (probability of not being readmitted)
Performance:
• The area under the ROC curve is 0.74.