Franckowiak et al reported a hospital all-cause thirty-day readmission index (HATRIX). This can help to identify a patient who may benefit from interventions to reduce admissions. The authors are from Wayne State University.
Patient selection: patient being discharged from the hospital
Parameters:
(1) number of previous admissions
(2) therapy with antiarrhythmic agents
(3) therapy with anxiolytic, sedative or hypnotic
(4) non-hypertensive congestive heart failure (CHF) in previous year
(5) essential hypertension
(6) chronic kidney disease
(7) anemia and other deficiencies
(8) pulmonary heart disease
(9) liver disease
(10) duration of admission in days
Parameter |
Finding |
beta-Coefficient |
number of previous admissions |
0 |
0 |
|
1 |
0.536 |
|
2 |
0.872 |
|
>= 3 |
1.644 |
antiarrhythmic agent |
no |
0 |
|
yes |
0.064 |
anxiolytic, sedative, hypnotic |
no |
0 |
|
yes |
0.053 |
CHF |
no |
0 |
|
yes |
0.208 |
essential hypertension |
no |
0 |
|
yes |
0.156 |
chronic kidney disease |
no |
0 |
|
yes |
0.361 |
anemia or other deficiency |
no |
0 |
|
yes |
0.230 |
pulmonary heart disease |
no |
0 |
|
yes |
0.181 |
liver disease |
no |
0 |
|
yes |
0.151 |
duration of admission |
|
0.034 * (days) |
value of X =
= SUM(points for all of the beta-coefficients) - 2.725
probability of readmission =
= 1 / (1 + EXP((-1) X))
Performance:
• The area under the ROC curve is 0.72.