- Poster presentation
- Open Access
Modified Early Warning Scores: inaccurate summation or inaccurate assignment of score?
© BioMed Central Ltd. 2010
- Published: 1 March 2010
- Emergency Medicine
- Vital Sign
- Large Dataset
- Electronic System
- Prospective Observational Study
Modified Early Warning Scores (MEWS) have been utilised to improve the recognition and medical response to patient deterioration. Their usefulness is, in part, dependent upon their accuracy and to date only small datasets have been analysed to determine the accuracy of MEWS . The aim of our study was to determine the accuracy of MEWS and the reasons for any inaccuracies found in a large dataset.
A prospective observational study of 9,672 vital sign sets and MEWS recordings in 315 adult medical and surgical patients admitted to four wards in both a tertiary and a metropolitan hospital over a 4-month period. Individual vital sign MEWS and total MEWS was assessed for accuracy against a computer-generated individual vital sign MEWS and total MEWS using the vital sign dataset and a modified MEWS algorithm.
Of the 9,672 total MEWS recorded, 3,504 (3,504/9,672, 36%) had discrepancies between the nurse-recorded and computer-generated total MEWS. Of these, 3,029 (3,029/3,504, 86%) underestimated the computer-generated total MEWS. In the majority of the total MEWS inaccuracies (2,443/3,504, 69.7%), the summation of the total MEWS from the individual MEWS was correct but the individual vital sign MEWS assignment was inaccurate. In 711 (711/3,504, 20.3%) cases the individual vital sign MEWS assignment was correct but the summation for a total MEWS was incorrect, and in 350 (350/3,504, 10%) cases both the individual vital sign MEWS assignment and summation for a total MEWS was incorrect.
The underscoring between nurse-recorded and computer-generated MEWS indicates that the paper-based MEWS system is less likely to trigger a timely medical review and appropriate treatment of a deteriorating patient. Further education and regular auditing on the assignment of the individual vital sign MEWS may improve the accuracy of the score given that the summation of the score was not the predominant issue. Equally, an electronic system to calculate and summate the individual vital sign MEWS may reduce the total MEWS error.