Display Settings:


Send to:

Choose Destination
See comment in PubMed Commons below
Ital J Gastroenterol Hepatol. 1997 Dec;29(6):495-500.

A validated dyspepsia symptom score.

Author information

  • 1Dept. of Gastroenterology, Meath/Adelaide Hospitals, Trinity College, Dublin, Ireland.



The primary endpoint in clinical trials involving patients with non-ulcer dyspepsia is subjective, i.e., reduction in symptomatology. Three attributes, reproducibility, responsiveness and validity, are necessary for the use of a symptom scoring system in a clinical trial.


The four most common symptoms in 50 dyspeptic patients were determined. To check the reproducibility of the symptom score, 48 patients and thirty control subjects were interviewed on two occasions (T0, T1), prior to any diagnostic or therapeutic intervention. Responsiveness was assessed by comparing the symptom scores of patients before (T0) and after (T2) treatment. Validity was assessed by comparing the symptom scores of dyspeptic patients to those of healthy volunteer subjects.


Reproducibility The median T0 (16.00 and 6.90) and T1 (15.00 and 6.56) scores of the patients and controls did not significantly vary during the one-week interval. Responsiveness The symptom scores at T0 had decreased at T2 for patients with gastro-oesophageal reflux (17.00 to 11.50, p = 0.0014), non-ulcer dyspepsia where Helicobacter pylori was eradicated (16.00 to 7.00 p = 0.0014), and duodenal ulceration (18.50 to 7.50, p = 0.0117) while there was an insignificant decrease (18.00 to 13.00, p = 0.0642) in non-ulcer dyspepsia patients who received a prokinetic agent. Validity The mean rank symptom score of 74 patients (71.74) was significantly higher than that of the control population (26.83), (p = 0.0001). The mean time taken to perform the questionnaire was 3.6 minutes.


This questionnaire is suitable for the assessment of symptoms in patients with dyspepsia.

Comment in

[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Loading ...
    Write to the Help Desk