Reliability of indicators of nursing care quality: testing interexaminer agreement and reliability

OBJECTIVE: this study sought to test the interexaminer agreement and reliability of 15 indicators of nursing care quality. METHODS: this was a quantitative, methodological, experimental, and applied study conducted at a large, tertiary, public teaching hospital in the state of Paraná. For data analysis, the Kappa (k) statistic was applied to the categorical variables - indicators 1 to 11 and 15 - and the interclass correlation coefficient (ICC) to the continuous variables - indicators 12, 13, and 14, with the corresponding 95% confidence intervals. The categorical data were analyzed using the Lee software, elaborated by the Laboratory of Epidemiology and Statistics of Dante Pazzanese Institute of Cardiology - Brazil, and the continuous data were assessed using BioEstat 5.0. RESULTS: the k-statistic results indicated excellent agreement, which was statistically significant, and the values of the ICC denoted excellent and statistically significant reproducibility/agreement relative to the investigated indicators. CONCLUSION: the investigated indicators exhibited excellent reliability and reproducibility, thus showing that it is possible to formulate valid and reliable assessment instruments for the management of nursing care.