Description

Pacala et al identified comorbid conditions that could predict the probability of repeated hospital admission in an elderly patient. This can help identify those patients who may require closer monitoring or more aggressive therapy. The authors are from the University of Minnesota.

Parameters:

(1) self-rated health

(2) coronary artery disease (including angina pectoris, myocardial infarction)

(3) diabetes mellitus in past year

(4) hospitalization in past year

(5) number of doctor visits in past year

(6) presence of an informal caregiver able to care for the patient for several days

(7) age

(8) gender

Parameter

Finding

Points

self rated health

poor

0.770

fair

0.552

good

0.340

very good

0.327

coronary artery disease

absent

0

present

0.390

diabetes mellitus in past year

absent

0

present

0.319

hospitalized within past year

no

0

yes

0.545

number of doctor visits in past year

<= 6 visits

0

> 6 visits

0.318

informal caregiver available (friend, relative, neighbor)

none

-0.738

present

0

age in years

< 75 years

0

75 – 79 years

0.255

80 – 84 years

0.327

>= 85 years

0.559

gender

male

0.257

female

0

where:

• The points assigned are the regression coefficients for each parameter from Appendix B.

• I am confused by the point assignment for informal caregiver. I would think the presence of a friend, relative or neighbor able to care for the patient would reduce the hospital admissions. However, as the table in Appendix B is written, an informal caregiver increases the risk of admission. Perhaps a person without a caregiver dies without being admitted.

X =

= SUM(points for all 8 parameters) – 1.802

probability of being repeatedly admitted to the hospital within the next 4 years =

= EXP(X) / (1 + EXP(X))