Why It Matters
"Retrospective data tells us about the past. Is it indicative of today’s performance? Does it represent the status of the unit today? We can’t answer those questions with old data."
SIGN UP FOR IHI EMAILS
Processing ...

Using Retrospective Data Alone Will Not Make Care Safer

By Carol Haraden | Friday, February 12, 2016

Improvement teams generally look at retrospective measurement data to understand performance, but rarely do they look at today’s data

Collecting data for measurement can be like driving while looking in the rearview mirror. Improvement teams generally look at retrospective measurement data to understand performance, but rarely do they look at today’s data. Data for monitoring, on the other hand, helps us understand the risks and threats right in front of us. IHI Vice President Carol Haraden explains the difference between real-time data for monitoring and retrospective measurement data for understanding trends and improvement — both of which are critical for improving patient safety.

How do you know your patients are safe?

If you walk onto your care unit or into your clinic today, how would you find out if your team is providing the best care possible for patients? You would look at the data, right? But what kind of data would you review? Data for measurement? Or data for monitoring?

We typically collect measurement data to gauge improvement in quality and patient safety. We track how many infections we had or how many patients fell. We might assemble this data in a report to tell us what happened this past month or this past quarter. We rarely look at today’s data; instead, teams generally look back at data for a prior time period to help them understand performance.

On the other hand, we collect monitoring data to understand what information we need right now as well as to manage a situation in the future. When you walk onto a patient care unit on any given day, the data at that point in time is different than the data from the night before. Patients’ conditions change, staffing changes, and the environment changes all the time. Data for monitoring helps us understand — in real time — the risks and threats right in front of us, and helps us develop a plan to mitigate or prevent those risks today.

For example, if you care for a cardiac patient on an EKG monitor, you want to know how that patient is doing right now. You don’t get that information by reviewing yesterday’s EKG readings. You want to see today’s strip. Previous readings are, of course, informative and enable you to see how the patient’s heart is responding to various stresses and medication. If you want to know the patient’s current condition, however, you have to look at the EKG data right now.

Monitor Data to Evaluate Current Risks

Huddles are the typical mechanism teams use to gather data for monitoring. For example, for a unit-level huddle, the team comes together in the morning to ask a series of questions to assess the actual and potential risks on the unit that day. Assessing staffing (including staff attending meetings off the unit for all or part of the day), equipment, severity of patient illness, events that might occur, and staff or patients’ concerns can lead to a shared level of understanding and situational awareness that increases preparedness. It’s not enough to just discuss the risks; the team must develop plans to prevent and mitigate risks.

A medical-surgical unit, a dermatology clinic, and a pediatrics unit all have different risks and issues. During a huddle on a children’s unit, the team might ask, “Are there any family issues today that might cause us concern? If so, how will we manage those?” For example, if there’s a custodial dispute between the parents of one of the patients, then the team needs to be aware of the potential for a problematic situation. The team may decide that asking the parents about their willingness to schedule their visits during different times is one potential mitigation strategy. The team may consult with social work and security. The goal of such huddles are to anticipate potential issues and develop a plan ahead of time to address them, should they occur.

Pressure ulcers are another common care issue for which teams are collecting measurement data. Your team is hopefully collecting that data daily, perhaps asking at the end of each day, “Did we have any pressure ulcers today?” While the care team’s goal is to prevent pressure ulcers, when they review that day’s data in retrospect, those pressure ulcers have already happened. What might the team do differently if they are using data for monitoring instead? The team might ask in the morning huddle, “Do we know of anyone at risk today for pressure ulcers? If so, what’s our prevention plan?”

Retrospective data tells us about the past. Is it indicative of today’s performance? Does it represent the status of the unit today? We can’t answer those questions with old data. We cannot manage a dynamic situation with data collected weeks ago. You might compare it to driving while looking in the rearview mirror.

Data Is Only Useful If You Use It

I’m not saying that data for measurement isn’t important. We need to both measure and monitor because we can’t manage everything moment to moment. Measurement helps us learn things we may not have realized without looking at the data over time.

Both real-time data for monitoring and retrospective data for understanding trends and improvement are critical. We can only maximize patient safety efforts through identifying the best ways to gather and use both types of data. No data makes anything safer for patients and staff unless you know how to use it correctly to prevent and mitigate harm and optimize care.

You may also be interested in:
Science of Improvement: Establishing Measures
On Demand Video: Building Skills in Data Collection and Understanding Variation

first last

Average Content Rating
(0 user)
Please login to rate or comment on this content.
User Comments