Collecting Data and Devising Metrics
Greg Maynard, Michelle Magee, and Jeff Schnipper
Introduction
Data collection, analysis, and presentation are key to the success of any hospital glycemic control initiative, enabling the glycemic management team to track improvements in processes and outcomes for hospital glycemic control, to make necessary changes to their quality improvement efforts, and to provide administrative personnel with financial justification for their time and labor. This section discusses the underlying key principles of data collection, analysis, and reporting. The section on collecting data and devising metrics presents an overview of this rapidly evolving field as compiled from a number of groups actively working in this area and the relatively few published reports of this work from the medical literature.
It is the intent of the SHM Inpatient Glycemic Control Task Force to provide a practical approach to data collection and measurement of the quality of inpatient glycemic control and to provide guidance for more uniform reporting of glycemic control metrics in the literature. A case is made for defining key indicators from each of three core domains: (A) targeted measures of glycemic control, (B) safety measures for hypoglycemia and extreme hyperglycemia, and (C) insulin utilization data.
Underlying Key Principles of Data Collection and Reporting
General considerations
- Prioritize what you collect. Don’t be data rich, info poor (a DRIP).
- To guide the performance improvement process, it is essential that the glycemic control team track performance longitudinally using a standard set of metrics.
At a minimum, core data should be collected on glycemic control, safety (including hypoglycemia), and insulin use patterns.
- Measuring outcomes is important, but focusing on performance indicators is essential in order to get quick feedback and will allow you to focus on the steps that lead to improved outcomes.
- Sampling/paper collection is quite acceptable if automated data collection is not yet possible. Collect just enough data to inform your team of baseline processes and clinical performance indicators and whether you are making a difference.
- Carefully define what you want to see. Imagine the end product of data collection and reporting and make sure it’s what you want.
- Define how data will be collected and reported and assign responsibility for carrying this out.
- Try different methods and measures — they will evolve over time.
Glycemic data
- Automated data collection (and reporting) is preferable whenever possible.
- Blood glucose (BG) readings should be collected and available at the point of care in a timely manner.
- Ideally, bedside glucose meter readings should be downloaded into a central database that interfaces with hospital/system main data repositories so that the data can be analyzed in conjunction with patient, service and unit data.
- This rapidly generates LARGE amounts of data, so the way you interpret and report results needs a lot of thought.
- Service-floor- and health-care-provider-specific data are helpful.
- Data collection and analysis should be separated for critical care and non–critical care units because the processes and goals of care are distinct to each of these care settings.
|
