Carga de trabalho e dimensionamento de recursos humanos de enfermagem em unidade de terapia intensiva pediátrica estimada pelos instrumentos NAS, TISS-28 e NEMS

Abstract

Objectives: Identifying, among the instruments Nursing Activities Score (NAS), Therapeutic Intervention Scoring System-28 (TISS-28), and Nine Equivalents of Nursing Manpower Use Score (NEMS), what best reflects nursing workload in a pediatric intensive care unit (PICU) and size up nursing human resources with this score; characterize the patients admitted to the PICU according to clinical and personal data; categorize and analyze the care-related items, grouped, in basic activities and therapeutic interventions; verify the correlation and concordance between the instruments and the influence of companion and case severity on nursing workload. Methods: Observational prospective cohort study with a quantitative approach conducted at the PICU in the São Lucas Hospital of the Pontifical Catholic University of Rio Grande do Sul (PUCRS). The sample consisted of all children admitted to the PICU from June 1, 2011, to June 31, 2012. For patient characterization, clinical and personal data were collected and the score Paediatric Index of Mortality 2 (PIM 2) was determined. In order to assess nursing workload, the instruments TISS-28, NEMS, and NAS were used, converted into working hours. Also, these scores were stratified into two groups, in Group 1 the items that comprise the category basic activities were gathered; and Group 2 gathered the items that comprise the other categories of ventilatory, cardiovascular, renal, neurological, and metabolic support, and specific interventions. Companion’s presence was assessed by work shift and then computed in 24-hour care. Data analysis was based on descriptive statistics, Student’s t-test, or ANOVA, Pearson’s correlation, simple linear regression, and the Bland and Altman model. Mortality was analyzed through the Standardized Mortality Ratio (SMR) and the Area Under Curve Receiver Operating Characteristic (AUROC) was used. The nursing personnel was sized up having a resolution from the Brazilian Federal Nursing Council (COFEN) as a basis. Results: The sample consisted of 490 hospitalizations, which totaled 4,617 observations. As for children characterization, 53% were men and there was predominance of infants and preschoolers, 39% were under 1 year of age, and 28% were in the age group between 1 and 5 years old. Regarding the origin, 44% came from the surgical center, 21% from the emergency room, 20% from a hospital ward and 16% from another hospital. Many children had more than one organ dysfunction during hospitalization, highlighting the fact that 61% had respiratory dysfunction. When assessing the average working hours of the nursing team estimated by the scores, we observed that the instrument NAS measured 13.6 hours and it was statistically superior, compared to TISS-28 and NEMS – 11.7 and 12.9 hours, respectively. In the analysis of average working hours stratified in the groups, we found that in Group 1 the NAS measured 10.2±2.5 hours, standing out in this assessment; however, when considering Group 2, the TISS-28 estimated 6.3±3.1 hours, standing out among the instruments. Good correlations were found between NAS and TISS-28, NAS (2) and TISS-28 (2), NAS (2) and NEMS (2), NEMS and TISS-28, and NEMS (2) and TISS-28 (2). When analyzing concordance using the Bland and Altman method, we observed that TISS-28 and NEMS were the instruments that showed graphically the best concordance. According to the NAS average value, nursing workload was higher with children from the emergency service (14.15±3.89 hours), without companion (14.13±2.42 hours), and there was a tendency to increase the workload as the amount of organ dysfunction increases. The NAS also proved to be effective to estimate case severity, revealing on the day of admission an adequate capacity to discriminate mortality [AUROC of 0.811 (CI 95% 0.726-0.896)], and compared to IMP 2 [AUROC of 0.813 (CI 95% 0.716-0.910)]. Personnel sizing using the workload estimated by the NAS found out the need of 36 nursing professionals to work in this unit. Conclusion: The NAS showed to be the instrument that best measured nursing workload and, using the hours estimated through this score, it was possible to size up nursing human resources at the PICU. As for the characterization of children admitted to the PICU within the study period, there was predominance of men, infants under 1 year of age, and children from the surgical center. This study grouped the items that comprise the instruments NAS, TISS-28, and NEMS, separating the category basic activities from the other categories of therapeutic interventions. The NAS stood out in Group 1 in the category basic activities and TISS-28 in Group 2, which gathered the other categories. The instruments TISS-28 and NEMS were those showing the best correlation and concordance. Through the analyses of total mean workloads, the NAS stood out to estimate the demand for nursing work in this hospital unit. There was an increased nursing workload when the child was without her/his companion. The NAS showed capacity to discriminate mortality and assess case severity.