A paired-comparison intervention to improve quality of medical records

Background: To evaluate the quality of medical record (MR) compilation in the Teaching Hospital of the Second University of Naples, Italy, after a controlled intervention for quality improvement. Methods: From the 66 wards of the Teaching Hospital, we selected eight homogeneous pairs of wards, matched for similar typology. For each pair, we randomized a ward to undergo a training course about correct compilation of MRs (treated group) and considered the remaining ward as a control (untreated group). For each section of MR we evaluated completeness and clarity of handwriting and presence and clarity of signature. Results: In general, the worst result in both groups was the absence of signature in the daily diary (76.6% in the treated group and 94.4% in the untreated group). The greatest differences between the two groups were detected in the compilation of the daily diary (absent/incomplete in 1.9% of the treated group compared with 21.9% of the untreated group; relative risk [RR] = 11, 95% confidence interval [CI] = 5.1-26.4) and the physical examination section (absent/incomplete in 2.8% of the treated group compared with 21.3% of the untreated group; RR = 7.5; 95% CI = 3.8-14.8). Conclusions: Comparison between the treated and untreated groups shows that there is a significant improvement in compilation of several sections of the MRs in the treated group. However, the results obtained were only partially satisfactory because of the poor quality of MR compilation in both groups.