Objective: To evaluate the impact of a programme of training, education and awareness on the quality of the data collected through discharge abstracts.
Study design: Three random samples of hospital discharge abstracts relating to three different periods were studied. Quality control to evaluate the impact of systematic training and education activities was performed by checking the quality of abstracting medical records.
Setting: The study was carried out at the Istituto Dermopatico dell'Immacolata, a research hospital in Rome, Italy; it has 335 beds specializing in dermatology and vascular surgery.
Measures: Error rates in discharge abstracts were subdivided into six categories: wrong selection of the principal diagnosis (type A); low specificity of the principal diagnosis (type B); incomplete reporting of secondary diagnoses (type C); wrong selection of the principal procedure (type D); low specificity of the principal procedure (type E); incomplete reporting of procedures (type F). A specific rate of errors modifying classification in diagnosis related groups was then estimated.
Results: Error types A, B and F dropped from 8.5% to 1.3%, from 15.8% to 1.6% and from 22% to 2.6% respectively. Error type D and E were zero in the third period of analysis (September-October 1997) compared with a rate of 0.7% and 4.1% in the third quarter of 1994. Error type C showed a slight decrease from 31.8% in 1994 to 27.2% in 1997. All differences in error types except incomplete reporting of secondary diagnoses were statistically significant. Five and a half per cent of cases were assigned to a different diagnoses related group after re-abstracting in 1997 as compared to 24.3% in the third quarter of 1994 and 23.8% in the first quarter of 1995.
Discussion: Training and continuous monitoring, and feedback of information to departments have proved to be successful in improving the quality of abstracting information at patient level from the medical record. The effort to increase administrative data quality at hospital level will facilitate the use of those data sets for internal quality management activities and for population-based quality of care studies.