An educational intervention that combined the dissemination of a clinical pathway with external auditing geared to reflection upon practice was effective in promoting changes in doctors' behaviours. The persistence of non-recommended practices and the variation among hospitals indicate that multiple factors influence performance and affect the effectiveness of interventions.

Two calibrated grip dynamometers were used to measure left and right handgrips of seven healthy volunteers. Ten measurements were made with each instrument on each hand. The results showed that the cheaper instrument read high by up to 5 kgf and that the more expensive instrument was accurate to within 1 kgf throughout its range. The results expressed as fractions of the mean for a set showed a broadly normal distribution with the reproducibilities of right and left hand results for the two instruments for a given individual being typical k 2 kgf. A conclusion of the study is that for statistical reasons caution should be exercised in interpreting the effects of a treatment or the progress of a disease from a very limited number of handgrip measurements. It is suggested that the mean of four measurements is a realistic compromise between practicability and accuracy.

No abstract