King's College London

Research portal

Early diagnostic suggestions improve accuracy of family physicians: a randomized controlled trial in Greece

Research output: Contribution to journalArticle

Olga Kostopoulou, Christos Lionis, Agapi Angelaki, Salma Ayis, Stevo Durbaba, Brendan Delaney

Original languageEnglish
Pages (from-to)323-328
Number of pages6
JournalFamily Practice
Issue number3
Publication statusPublished - 3 Jun 2015

King's Authors


Background. In a recent randomized controlled trial, providing UK family physicians with ‘early support’ (possible diagnoses to consider before any information gathering) was associated with diagnosing hypothetical patients on computer more accurately than control. Another group of physicians, who gathered information, gave a diagnosis, and subsequently received a list of possible diagnoses to consider (‘late support’), were no more accurate than control, despite being able to change their initial diagnoses.
Objective. To replicate the UK study findings in another country with a different primary health care system.
Methods. All study materials were translated into Greek. Greek family physicians were randomly allocated to one of three groups: control, early support and late support. Participants saw nine scenarios in random order. After reading some information about the patient and the reason for encounter, they requested more information to diagnose. The main outcome measure was diagnostic accuracy.
Results. One hundred fifty Greek family physicians participated.The early support group was more accurate than control [odds ratio (OR): 1.67 (1.21–2.31)]. Like their UK counterparts, physicians in the late support group rarely changed their initial diagnoses after receiving support. The pooled OR for the early support versus control comparison from the meta-analysis of the UK and Greek data was 1.40 (1.13–1.67).
Conclusion. Using the same methodology with a different sample of family physicians in a different country, we found that suggesting diagnoses to consider before physicians start gathering information was associated with more accurate diagnoses. This constitutes further supportive evidence of a generalizable effect of early support.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454