Measurement should speed improvement, not slow it down. Often, organizations get bogged down in measurement, and delay making changes until they have collected all of the data they believe they require. Remember, measurement is not the goal; improvement is the goal. In order to move forward to the next step, a team needs just enough data to make a sensible judgment as to next steps.
Sampling is a simple, efficient way to help a team understand how a system is performing. In cardiac surgery, the patient volume is typically low enough to allow tracking of key measures on all patients. In other settings, however, sampling can save time and resources while accurately tracking performance.
A Valuable Lesson
The Sunnybrook Health Sciences Centre in Toronto, Ontario, Canada, learned this lesson when they participated in the Institute for Healthcare Improvement's (IHI) Improving Asthma Care Breakthrough Series Collaborative in 1995. The team sought some baseline data regarding the age distribution of patients visiting the Emergency Department (ED) for asthma care. So, they put in a formal request to their Information Systems (IS) department for a report with this information. And then they waited. Impatient that the request was taking too long, one team member, an emergency doctor, decided to take the matter into his own hands, literally. This physician went through the charts for all ED patients in a three-month period, and manually noted the ages of asthma patients. His informal research provided a sampling of 94 patients, with an age distribution as shown in the graphs below.
The team had enough confidence in this data that it decided to proceed with its improvement agenda. Sure enough, when the IS department finally delivered the report nearly two months later, it confirmed the team's instincts. With a sample more than four times the informal sample taken two months earlier, the results were nearly identical:
This team would have learned nothing more by waiting for more data. They saved valuable time by sampling and depending on their own sound judgment to know when enough data was enough.
