Development of Dashboards to Provide Feedback to Home Care Nurses

An increasing number of Americans receive care from home health agencies (HHAs), a number that is projected to grow as the U.S. population ages. HHAs in the U.S. typically provide care to patients who have been discharged from a hospital, where they may have received skilled nursing care, rehabilitation therapies, and social work support. The quality and safety of care has been the focus of a number of quality improvement strategies in healthcare organizations, including in HHAs.

This study focused on providing feedback on quality performance measures related to the care of patients with chronic heart failure (CHF) to try and reduce avoidable hospital readmissions, seen as a metric of poor care quality. While some re-hospitalization risk factors, such as race and gender, cannot be controlled by clinicians, other risk factors are amenable to more intensive monitoring, such as medication reviews and regular assessments of weight gain, nutritional status, and functional capacity. This project designed and conducted a usability evaluation of health information technology dashboards that use visualization techniques to provide feedback on metrics for CHF patients to improve patient care.

The specific aims of the project were as follows:

• Operationalize quality performance metrics related to the management of patients with CHF that can be used by home care nurses to improve their practice. 
• Provide feedback to home care nurses, which maximizes their comprehension of the data. 
• Develop a set of flexible, dynamic quality dashboards individualized to nurse user preferences. 

Focus groups were conducted with nurses to identify existing and actionable quality performance measures related to CHF patient care and those relevant to home care nurses. They also collected data via an online survey to explore the impact of numeracy skills and graph literacy of home care nurses on their comprehension of visualized data. Researchers developed interactive web-based prototype dashboards for monitoring of patient vital signs and included basic decision support to indicate when a patient’s weight gain or vital signs were outside guideline recommendations or physician orders. The dashboards were evaluated for function, user satisfaction, and usability.

Researchers found that nurses prioritized information about a patient’s signs and symptoms, specifically weight gain, as an area of practice where they would like to see trends over time and receive real-time feedback. Individual differences in the ability to understand information presented in different graphical formats were identified: data in the bar graph format was associated with the highest comprehension levels. The dashboard prototype had high usability and was evaluated positively by nurse users. Overall study findings highlight the importance of understanding user information needs, and individual variation in user characteristics when designing technology such as dashboards.