Reliability of symptom assessment in dyspepsia

To examine the variation between doctors in their clinical assessment of dyspeptic symptoms.

Simulated presentation of three dyspeptic symptom complexes to a total of 27 general practitioners, who each interviewed the 'patient' in accord with their own clinical practice A structured record form was used to record details of the history elicited.

Simulation of a primary care consultation.

Although overall agreement in symptom assessment was reasonable for each of the three symptom complexes, both the accuracy and consistency of assessment varied considerably for individual symptoms. Inconsistency and inaccuracy in the assessment of epigastric pain were unexpectedly large, demonstrating that neither the nature nor severity of pain were communicated reliably.

The findings illustrate the feasibility of quantitative appraisal of patient-doctor communication in respect of symptoms and show that there is much variation in the reliability with which different symptoms are communicated. Difficulty in characterizing abdominal pain reliably may contribute to the well recognized inadequacy of a conventional clinical history in identifying the cause of a patient's dyspepsia.