Description

Kruis et al developed a score for separating patients with Irritable Bowel Syndrome (IBS) from those with organic bowel disease. This uses symptom information supplied by the patient with physical and laboratory findings supplied by the physician. The score was developed at the University of Munich in Germany.


 

Components:

(1) Patient questions (after Figure 1, page 3): 7 items

(2) physician checklist

 

Physician Checklist

Parameter

Finding

Points

abnormal physical findings and/or history pathognomonic for any diagnosis other than IBS

present

-47

 

absent

0

erythrocyte sedimentation rate

> 20 mm in 2 hours

-13

 

<= 20 mm in 2 hours

0

WBC count

leukocytosis with > 10,000 per µL

-50

 

<= 10,000 per µL

0

hemoglobin and gender

< 14 g/dL and male

-98

 

>= 14 g/dL and male

0

 

< 12 g/dL and female

-98

 

>= 12 g/dL and female

0

history of blood in stools

yes

-98

 

no

0

after Figure 1, page 3

 

score =

= (points for patient questionnaire) + (points for physician checklist)

 

Interpretation:

• maximum score: 87

• minimum score: -404

• A cutoff score of 44, using >= 44 as indicating IBS, had a positive predictive value of 94% when the prevalence of IBS is 20%.

• Based on the data in Figure 2, page 5, scores >= 44 were very likely to be IBS; scores 20-43 were likely to be IBS; scores of 0 to 19 were likely to be organic bowel disease; scores < 0 were very likely to be organic bowel disease.

 

Prevalence of IBS

Positive Predictive Value for score >= 44 (percent with IBS)

Negative Predictive Value for score < 44 (percent with organic bowel disease)

10%

87.7%

96.1%

20%

94%

 

30%

96.4%

86.5%

50%

98.4%

73.3%

70%

99.3%

54.0%

90%

99.8%

23.4%

 

The positive predictive value can be represented by a 5th order polynomial.

 

negative predictive value =

= (-0.008589 * ((prevalence as percent from 0 to 100)^2)) - (0.030571 * (prevalence as percent from 0 to 100)) + 96.533214

 


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