Measure to Improve

As a central theme of practice redesign, measurement has been consistently associated with practice improvements. Improvement doesn't happen without measurement. Numerous distinguished health organizations, including the Dartmouth-Hitchcock Health System, Mayo Clinic, the University of Rochester and others have begun using clinical instrument panels as a formal means on measurement in daily practice.

A clinical instrument panel is a balanced set of performance measures that reflect the inputs, processes and outcomes of your practice integrated into daily practice as a means of (a) diagnosing weaknesses and strengths, (b) improving and innovating health care delivery processes, (c) managing patients and the practice, and (d) evaluating changes over time.

A proactive, systematic and daily strategy of measuring the inputs, processes and outcomes of practice will inform practice level decision-making both for patients and for the practice. It will help you understand how satisfied your patients are with your care, how satisfied your staff is with working in your practice, the outcomes you are achieving with your practice patterns, and the costs and revenues that are associated with how you practice.

1. Define AIMS for measurement. Aims for measurement might include reporting to health system, to assist strategic planning of the practice, to identify strengths and opportunities for improvement in the practice, to monitor how well the practice is doing in areas in which it seeks improvement. Aims of measurement might include patient satisfaction and loyalty, staff morale and loyalty, clinical outcomes both for acute and chronic conditions, appropriate use of resources, financial performance. Aims should be defined specifically for the following SMART characteristics:

S - Specific
M - Measurable
A - Attainable
R - Relevant
T - Timely

2. Select a balanced set of measures. Select a set of measures that are balanced, reflecting the practices inputs, processes, and outcomes. A minimum set of measures is 8-10 measures that are collected on regular basis. This set should be dynamic and changed when the aims of the practice evolve and change. Write down operational definitions. (see linked 'all star performance metrics' and the 'clinical instrument panel'. In thinking about outcomes, consider the 'clinical value compass' approach which balances the types of outcomes (clinical, cost, satisfaction, functional status).

3. Design how measures will be graphed or displayed. Run charts are the most informative of displays with the value of the measure on the 'Y' axis and the time period on the 'X" axis. Bar Charts are also helpful in showing changes, especially for comparisons. Below are two mock graphs for run and bar charts. Plan for how measures will be displayed and communicated. Put up a "Data Wall' which is a bulletin board in an area where all staff have a chance to view the data. Arrange the graphs to provide a complete picture of improvements happening in the practice (see Data Wall link). Consider also producing a 'flash-sheet' with all or selected graphs produced on a single sheet of paper and distributed to staff on a regular basis. (use the measurement assistant link to produce graphs and flash-sheets)

4. Define data collection methods and frequency. For each measure, write down in a protocol or table how it will be collected and analyzed, how often, and by whom. Data may be available from other sources (your health system's IS department). Don't duplicate data collection efforts. Data can be found in charts, administrative records, registration forms, surveys etc. Be clear how you are collecting data and communicate this to all staff.

5. Assign a measurement captain. Assign responsibility to an individual in the practice to supervise the collection, analysis and display of data. This person can be a nurse, secretary, provider or front office staff. Select someone preferably who has a background in data collection and use of data. If no one fits this qualification, consider sending this individual for training in measurement methods.

6. Build measurement into daily work flows. Design or adapt forms that assist easy collecting of data in the course of work. Example, a table for type of telephone call, a checklist for chart reviews, small cards for frequent patient satisfaction survey, a time card to be stamped in and out by patients for cycle time surveys. Also consider doing periodic larger data collection efforts such as surveys of patients, providers and staff. (see links in resources for patient viewpoint (pt satisfaction), team work life (staff satisfaction), and office manager surveys from the IHI Idealized Design of Clinical Office Practices as examples- feel free to modify and adapt to your own practice.

7. Revise as needed. Regularly assess whether your data collection efforts are answering the critical questions for your practice. If not, discard old measures and define and test new measures that better reflect the evolution and change of your practice. If improvement within a specific area is meeting practice goals or is not changing over long periods of time, rethink which measures make sense. Remember, information is data that makes a difference.

Hess AM, Nelson EC, Johnson DJ, Wasson JH. "Building and Idealized Measurement System to Improvement Clinical Office Practice Performance." Managed Care Quarterly. 1999; 7(3): 26-38.

Nelson EC, Splain ME, Batalden PB, Plume SK. "Building Measurement and Data Collection into Medical Practice." Annals of Internal Medicine. 128: 460-466.