Description

Tangiisuran et al reported the Brighton Adverse Drug Reactions Risk (BADRI) model for predicting the risk of an adverse drug reaction in a hospitalized patient. The authors are from University Sains (Malaysia), King's College London, Brighton and Sussex Medical School, Catholic University of the Sacred Heart (Rome), Ghent University Hospital, Erasmus University and Delft University of Technology.


Patient selection: older adult (median age 85 years)

 

Parameters:

(1) hyperlipidemia

(2) number of medications

(3) length of hospital stay in days

(4) use of anti-diabetic agent

(5) WBC cell count on admission

 

Parameter

Finding

Points

Beta-Coefficient

hyperlipidemia

absent

0

0

 

present

1

1.199

number of meds

< 8

0

0

 

>= 8

1

1.194

length of stay

< 12 days

0

0

 

>= 12 days

1

0.819

anti-diabetic drugs

no

0

0

 

yes

1

0.645

WBC count

not high

0

0

 

high

1

0.437

 

where:

The cutoff for a high white blood cell count was not stated.

 

total score =

= SUM(points for all 5 parameters)

 

value of X =

= SUM(all 5 beta-coefficients) - 3.628

 

probability of an adverse drug reaction =

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

 

Interpretation:

minimum score: 0

maximum score: 5

The higher the score the greater the risk of an adverse drug reaction.

The maximum Youden index is for a score > 1.

 

Performance:

The area under the ROC curve is 0.73.


To read more or access our algorithms and calculators, please log in or register.