OBJECTIVE: To assess the face validity of both simple and advanced quality indicators for prescribing in general practice. METHODS: In a three-round Delphi study, 100 randomly selected general practitioners (GPs) in Denmark rated 18 indicators for prescribing of non-steroidal anti-inflammatory drugs. All indicators were based on prescription register data and focused on different prescribing aspects. Advanced indicators contained information at the patient level, viz. age, sex and history of drug use, while simple indicators only used drug statistics at practice level. Indicators were rated on a nine-point Likert scale. Consensus among GPs was defined as interquartile ranges of three or less. A median rating of 7-9 was interpreted as face validity and a median rating of 1-3 as no face validity. RESULTS: Participation in the study was accepted by 44 GPs and 37 completed all three rounds. Three indicators based on patient level data and focusing on adverse effects were assessed to have face value. One indicator focusing on costs and based on practice level data was considered unsuitable for evaluating the quality of prescribing. Consensus was not reached for the remaining indicators. CONCLUSIONS: GPs do not regard simple indicators based on aggregated data at practice level as suitable for evaluating the prescribing quality in general practice, but prefer indicators that rest on clinical data at the patient level.
OBJECTIVE: To assess the face validity of both simple and advanced quality indicators for prescribing in general practice. METHODS: In a three-round Delphi study, 100 randomly selected general practitioners (GPs) in Denmark rated 18 indicators for prescribing of non-steroidal anti-inflammatory drugs. All indicators were based on prescription register data and focused on different prescribing aspects. Advanced indicators contained information at the patient level, viz. age, sex and history of drug use, while simple indicators only used drug statistics at practice level. Indicators were rated on a nine-point Likert scale. Consensus among GPs was defined as interquartile ranges of three or less. A median rating of 7-9 was interpreted as face validity and a median rating of 1-3 as no face validity. RESULTS: Participation in the study was accepted by 44 GPs and 37 completed all three rounds. Three indicators based on patient level data and focusing on adverse effects were assessed to have face value. One indicator focusing on costs and based on practice level data was considered unsuitable for evaluating the quality of prescribing. Consensus was not reached for the remaining indicators. CONCLUSIONS: GPs do not regard simple indicators based on aggregated data at practice level as suitable for evaluating the prescribing quality in general practice, but prefer indicators that rest on clinical data at the patient level.
Authors: N D Yeomans; Z Tulassay; L Juhász; I Rácz; J M Howard; C J van Rensburg; A J Swannell; C J Hawkey Journal: N Engl J Med Date: 1998-03-12 Impact factor: 91.245
Authors: C J Hawkey; J A Karrasch; L Szczepañski; D G Walker; A Barkun; A J Swannell; N D Yeomans Journal: N Engl J Med Date: 1998-03-12 Impact factor: 91.245
Authors: S C Lewis; M J S Langman; Joan-Ramon Laporte; John N S Matthews; Michael D Rawlins; Bengt-Erik Wiholm Journal: Br J Clin Pharmacol Date: 2002-09 Impact factor: 4.335
Authors: Sture Rognstad; Mette Brekke; Arne Fetveit; Olav Spigset; Torgeir Bruun Wyller; Jørund Straand Journal: Scand J Prim Health Care Date: 2009 Impact factor: 2.581