| Summary: | Thesis (M.Sc.)--Memorial University of Newfoundland, 1997. Medicine Bibliography: leaves 82-89 Purpose: The purpose of this study was to develop and validate a symptom-based outcome measure for clinical trials in nonulcer dyspepsia (NUD). -- Methods: Patients referred to the GI outpatient clinic with chronic upper abdominal pain were approached for this study. Participants with normal endoscopy who met the inclusion criteria were enrolled and classified into standard subgroups of NUD. The ulcerlike subgroup received acid suppressive therapy and the dysmotilitylike subgroup received prokinetic agents. -- Each subject completed a questionnaire at the initial visit (T1), one week later before treatment (T2), and after one month of treatment (T3). Each subject selected the symptom most important to them. The frequency and severity of the selected symptom was recorded as was a global assessment by the subject of their overall status. Other data recorded included a physician global assessment, the subject's and physician's impression of change in symptoms with treatment, and the subject's antacid use. -- Results: Forty-four subjects were enrolled. The primary outcome measure of the study was the product of the selected symptom's frequency and severity. Instrument reliability was assessed by Spearman rank correlation r=0.85 and the intraclass correlation coefficient = 0.83. -- Good correlation between the measure and patient global assessment (lrl=.596) and between the measure and patient assessment of treatment response (Irl=.584) was noted. The physician global assessment was moderately related to the measure (lrl=.437). Some relationship was seen between the measure and physician assessment of treatment response (lrl=.329). There was no relationship with change in antacid use (lrl=.143). -- Conclusion: The main measure was reliable in untreated subjects and responsive in subjects who responded to therapy. This measure also appeared to be valid. The method of combining frequency and severity of the subject-selected symptom into a product was shown to be a useful means of assessing treatment effect in NUD patients. This study documents a reliable, responsive, and valid new outcome measure for use in clinical trials of NUD.
|
|---|