Abdominal pain patient referrals to emergency surgical service: appropriateness of diagnosis and attitudes of general practitioners.

This study evaluate the need for general practitioners referrals and self referrals of acute abdominal pain patients to emergency surgical service, the appropriateness of GP referral diagnosis and their attitudes dealing with abdominal pain. In three months period all acute abdominal pain patient referrals to our hospital emergency surgical service were audited. Data on final diagnosis, surgical treatment, admission to hospital and surgery performance were recorded. Self referral or GP referral, referring GP diagnosis, referral letters indicating presenting complaint or history, axillar and rectal temperature measurement, laboratory checking and abdominal radiography checking by GP were recorded as well. Also, GPs examination details as palpation, auscultation and digit-rectal checking were recorded. We calculated sensitivity, specificity, positive and negative predictive value (PV) for referring diagnosis. Self referrals and GP referrals differences were evaluated. During the study 318 patients were admitted. A total of 163 (51.25%) referrals were deemed inappropriate; 102 (52.6% of GP referrals) and 61 (49.2% of self referred) (p < 0.05). There were no differences in general treatment, hospital admission and operative treatment in self referred and GP referred groups (p < 0.05 for all three categories). Sensitivity, specificity, positive and negative predictive values for most frequent GP referral diagnoses were: abdominal colic/abdomen in observation 0.78; 0.66; 0.74; 0.70; acute appendicitis 0.37; 0.92; 0.44; 0.90; acute abdomen/peritonitis 0.30; 0.97; 0.54; 0.92; constipation 0.95; 0.98; 0.85; 0.99; and ileus 0.83; 0.97; 0.50; 0.99. Data on GP including clinical examination, patient history and running basic diagnostics were poor. Our results suggest that a general agreement within the profession about what constitutes a necessary hospital referral is necessary. GP consultation quality must be improved by booking more time per patient and by giving more medical/technical attention to patients.