Evaluation of Process and Effects of Change

Principles for Using Data
to Support Clinical Improvement ¹

These eight principles provide guidelines for using data to support improvement in clinical settings:

1. Seek usefulness, not perfection, in measurement
   The utility of data is directly related to the timeliness of feedback and the appropriateness of its level of detail for the persons who use it. The goal is continuous improvement with concurrent, ongoing measurement of impact.
    
2. Use a balanced set of process, outcome, and cost measures
   Medical care systems are comprised of sub-elements that interact, flow into and out of one another, and contain feedback loops. They produce a fluid family of results that include: clinical outcomes, functional status, risk level, patient satisfaction, and costs.
    
   Balanced measures may cover "upstream" processes and "downstream" outcomes to link:
    
   • causes with effects
   • anticipated positive outcomes and potential adverse outcomes
   • results of interest to different stakeholders such as patient, family, employer, community, payer, and clinician - because participants have differing viewpoints on the relative importance of the many manifestations of care
   • cumulative results related to the overall aim
   • specific outcomes for a particular change cycle.²
   
    
3. Keep measurement simple (think big, but start small)
   Data collection must balance the desire for a rapid rate of improvement with an understanding of the complexity of the system. Strive to focus data collection on a limited, manageable, and meaningful set of measures.
    
4. Use quantitative and qualitative data
   Quantitative data on performance is used to create tension for change and to measure impact on clinical behavior. Qualitative data is used to learn how physicians, nurses, patients, or others experienced a new procedure or system.
    
5. Write down the operational definitions of measures
   The better the operational definition, the better the data elements. The better the data elements, the more reliable and valid the aggregate measures.
    
6. Measure small, representative samples
   Use a sampling strategy that avoids the costs and trouble of collecting data on all patients and on all encounters. The emphasis is on usefulness, not perfection.
    
   Pilot testing reduces the scope of unexpected effects and maximizes the opportunity to learn about the implementation and effects of a given change.³
    
7. Build measurement into daily work
   Every process emits data that can reveal how that process is performing. Information systems can accomplish most quality measurement if key data are captured in coded form during the provision of routine care.⁴
    
8. Develop a measurement team
   Avoid having one person do all of the data collection. Team up to lighten the workload, add knowledge, and boost morale.

-Return to the Top-
-Return Home-

Continue to Evaluation: Steps for Improving Clinical Practices