J D G Fearn-Smith1, P H Evans, G Harding, J L Campbell. 1. Peninsula Medical School (Primary Care) Smeall Building, St. Luke's Campus, Exeter, Devon EX1 2LU, UK. jason@fearn-smith.freeserve.co.uk
Abstract
AIMS: To develop and pilot a survey instrument assessing general practitioners' (GP) attitudes to the diagnosis and management of one form of pre-diabetes, impaired glucose tolerance (IGT) and to assess the performance of the questionnaire. METHODS: Qualitative data together with an audit were used to generate questionnaire items, which were then subjected to a process of pre-piloting and piloting to generate a finalised item list. The pilot questionnaire was sent to 222 principal GPs in three PCTs in South West England. The first fifty responders were asked to complete the questionnaire again 2 weeks later. Principal components analysis with a Varimax rotation was used to detect latent factors within the data that may help to explain the attitudes of GPs. RESULTS: The response rate after one reminder was 54.1%. Four robust factors were identified which were internally consistent (range of Cronbach's alpha=0.79-0.65), homogeneous (item-total correlations=0.60-0.21), and stable (test-retest correlation=0.74-0.58) accounting for 31.1% of the variance. The predictive validity of the item list was assessed (P=0.02 for factor 1). CONCLUSIONS: The PAtH questionnaire identifies four factors that help to describe GPs attitudes to the diagnosis and management of IGT.
AIMS: To develop and pilot a survey instrument assessing general practitioners' (GP) attitudes to the diagnosis and management of one form of pre-diabetes, impaired glucose tolerance (IGT) and to assess the performance of the questionnaire. METHODS: Qualitative data together with an audit were used to generate questionnaire items, which were then subjected to a process of pre-piloting and piloting to generate a finalised item list. The pilot questionnaire was sent to 222 principal GPs in three PCTs in South West England. The first fifty responders were asked to complete the questionnaire again 2 weeks later. Principal components analysis with a Varimax rotation was used to detect latent factors within the data that may help to explain the attitudes of GPs. RESULTS: The response rate after one reminder was 54.1%. Four robust factors were identified which were internally consistent (range of Cronbach's alpha=0.79-0.65), homogeneous (item-total correlations=0.60-0.21), and stable (test-retest correlation=0.74-0.58) accounting for 31.1% of the variance. The predictive validity of the item list was assessed (P=0.02 for factor 1). CONCLUSIONS: The PAtH questionnaire identifies four factors that help to describe GPs attitudes to the diagnosis and management of IGT.