Laboratory investigation utilization in pediatric out-patient department Ramathibodi Hospital

Laboratory investigations are inappropriately utilized in the Pediatric OPD of Ramathibodi Hospital and this is due to lack of knowledge and unawareness of the cost of tests.

1) To obtain information on laboratory utilization. 2) To evaluate the appropriateness of the residents' test ordering behavior. 3) To estimate the cost of unnecessary tests. 4) To reduce the utilization by combined education, feedback, and peer-review strategies.

Randomized controlled trial carried out from January 1987 to May 1988, consisting of pre-intervention period (Jan-May 1987). intervention period (June-November 1987), and post-intervention period (July 1987-May 1988).

There were 3 intervention groups and 3 control control groups which were stratified by level of seniority as 10 first year residents (R1I vs R1C), 8 second year residents (R2I vs R2C), and 18 staff (SI vs SC).

For the residents, education on diagnostic tests, chart audit and feedback on their test ordering behavior were provided twice a week. The second year residents also played an active role in giving comments on the first year residents' performances. For the staff, the strategy used was peer-review. The simulated cases derived from the actual medical records serviced by the residents were reviewed and rated, according to the appropriateness of the residents' test ordering behavior, from 0 (poor) to 10 (excellent). A score of less than 5 was considered to indicate inappropriate ordering behavior.

1) Numbers and cost of tests per patient for each group. 2) The achievement of a score indicating appropriate test ordering behavior of the resident groups.

Kruskal-Wallis test, Mann Whitney test, Wilcoxon test and Chi-square test.

1. The utilization pattern was not correlated to the level of seniority of the physicians. 2. There was inappropriate laboratory utilization in this group of residents: overutilization defined as should not be ordered (26.9%), and underutilization defined as should have been ordered (17.7%). Most of the inappropriateness occurred in the ordering of microbiology (50%) and special blood chemistry (40%). 3. Only the number of tests per patient ordered by the R2I group was significantly lower than that of the R2C group in the late post-intervention period. The cost of tests per patient ordered by the R2C group was significantly increased as compared to their baseline. This suggested that the combination of education, chart audit and active feedback strategies was more effective than each strategy alone. 4. The appropriateness of the test ordering behavior of the resident intervention groups improved during the post-intervention period, although the change was not statistically significant. The control groups' behavior was statistically worse. This suggested that the strategy prevented the inappropriateness of test ordering behavior. 5. With the intervention, the overutilization significantly decreased while the underutilization increased. The underutilization might be a consequence of cost containment on laboratory utilization. 6. The cost of unnecessary tests accounted for 17.6 per cent of the total cost of tests ordered by the residents.