Objectives: In dyspepsia few data are available from the primary care setting on how selective, risk-factor-oriented endoscopy compares with mandatory endoscopy in the diagnostic outcome and in direct and secondary costs. We studied this in a two-armed multicentre trial (Omega-project) with primary care physicians. Material and methods: Patients were enrolled and treated by primary care physicians and referred to a gastroenterologist for upper gastrointestinal endoscopy (UGE). Patients were enrolled in the study if they had had epigastric complaints for more than 1 month and no obvious signs or history of organic disease. In the first arm of the study endoscopy was mandatory, in the second selective, i.e. according to a predefined risk profile. Patients enrolled were treated with prokinetic drugs for 2 months. A further indication for endoscopy was non-response to treatment (reduction of the initial symptoms score by less than two-thirds) in the study with selective endoscopy and relapse within the 2-month follow-up period in both studies. The direct costs from number of consultations with the primary care physician, UGEs, number of prescriptions per patient and also absenteeism in days per week were carefully registered in both groups. Results: All 172 patients of the mandatory endoscopy study and 203/656 patients enrolled in the selective endoscopy study had an UGE (125 at admission, 78 in the follow-up period). Patients were treated for 4 weeks (cisapride or domperidone) and thereafter followed for 8 weeks; at the end of the observation period the response rates were 80% and 79%, respectively. The prevalence of gastric cancers was similar in both groups (> 1%) but extrapolation from the data collected with compulsory endoscopy suggests that two-fifths of the anticipated peptic lesions remained undetected by following the selective strategy. The cost analysis revealed a 31% cost reduction with the selective strategy - in the Swiss cost system - through a reduction in the number of endoscopies by 67%. Conclusion: Selective UGE is cheaper and appears not to compromise the response to prokinetics, however, its diagnostic power is less than with mandatory UGE.
