The reliability and validity of work measurement in Australian general practice consultations.

OBJECTIVES

To examine the reliability of relative work value assessment in general practice consultations and to determine whether different methods of assessing work produce consistent rankings.

DESIGN

Cross-sectional observational assessment of general practice consulations.

SETTING

General practices in Victoria between October 1991 and October 1992.

PARTICIPANTS

686 patients attending one of 58 general practitioners (GPs) drawn from a random, stratified sample.

METHODS

Each participating GP had one day of consultations videotaped. They rated the work value of each consultation by using a magnitude estimation scale relative to a reference vignette. Three GP observers independently applied the same scale to the videotaped consultations. After three months, the observers applied a second measurement of work value, a compensation scale (also relative to the reference vignette), to the videotaped consultations. Duration of consultation was the third rating method.

MAIN OUTCOME MEASURES

The reliability of work value assessment for each scale. Consultation rank order correlation coefficients among all rating methods.

RESULTS

Observer reliability was high for both scales. Practising GPs showed lower levels of reliability in assessing the work value of their consultations. Strong positive correlations were found for consultation rankings among the observer scales and duration of the consultation. The duration of the consultation emerged as an important predictor of consultation work value.

CONCLUSIONS

Scaling methods appear to be of little value to the practising GP in reliably assessing the relative work value of their consultations; training in the use of these scales may improve their reliability. However, the duration of consultation may be a reasonable proxy for relative work value assessment in general practice consultations.