Abstract
PROCEnf-USP is a decision support system that offers staff nurses at the University Hospital in Sao Paulo, Brazil a set of questionnaires to guide patient assessment. Using the answers, the system computes probabilities for nursing diagnoses based on NANDA-I. This study aims to evaluate PROCenf in terms of its reliability (Does it consistently gives the right diagnose?) and its validity in Brazil (Does NANDA-I terminology represent the healthcare experience of a Brazilian population)
Description:
The Institute of Medicine compiled several reports that support the use of technology to improve the quality and the safety of the healthcare provided in the United States (2000, 20003, 2007, 2010). Computers are able to structure information and communication, becoming an intellectual instrument. In this context, decision support systems (DSS) are recognized as important tools, capable of processing high volumes of data and increasing productivity.
PROCEnf is a DSS. The system offers the staff nurse a set of questionnaires to guide the nurse when performing a patient assessment. The questionnaire has drop down menus with answers. Using these answers, the system computes probabilities for nursing diagnoses, and suggests a set of diagnoses that best fit the clinical scenario. The algorism is based on the North American Nursing Diagnosis Association – International (NANDA-I), which is being embraced worldwide. This study aims to evaluate PROCenf in terms of its reliability (Does it consistently gives the right diagnose?) and its validity in Brazil (Does NANDA-I terminology represent the healthcare experience of a Brazilian population)
This is a secondary data analysis of data collected as part of a larger, on-going project that supports the increase use of informatics solutions in nursing. PROCEnf was designed and implemented at the University Hospital of the University of São Paulo in 2009. Data were collected between January 1, 2011 and December 31, 2011 in the medical and surgical units will be analyses using Logistic regression. This study will calculate the effect of each answer choice selected in the assessment questionnaires on the suggested nursing diagnoses. The evaluation of these effects will inform nurses, nurse leaders, hospital administrators and engineers: a) if the software PROCEnf correctly mapped NANDA-I (reliability measure), and b) if NANDA-I accurately illustrates the clinical cases in Brazil (validity measure).