How do you build a new reporting system?

IF banner MARCH12 (jpg) EN

Giving health care facilities a tool that lets them analyze pan-Canadian data in real time is not a typical approach to database development. 

But when it came to medication incidents, that’s exactly what stakeholders asked for—and it’s what we delivered.

The National System for Incident Reporting (NSIR) lets facilities report medication and IV fluid incidents—which range from near misses to those causing harm to patients, including the wrong drug, dose, patient or time.

“The feedback has always been incredibly positive,” says Paul Sajan, a program lead on the project. “There’s been strong support from the patient safety community.”

The seed for the system was planted at a workshop in 2000. It came at a time when several international landmark studies had quantified lives lost and attached dollars to adverse events in hospitals. That spurred a desire to understand and improve medication safety in Canada.

There was a definite information gap, so in 2002, a group of national health care organizations developed a business plan for a medication incident reporting and prevention system and asked us to build it.

So where do you start when building a reporting system?

With funding secured from Health Canada and our small five-person team (plus IT) in place, we looked at what others had done and then brought 50 experts together in 2004 to gauge information needs and determine what the system needed to look like.

“One major piece of feedback we received was if you’re going to do this and you don’t protect the individuals involved, it’s going to be hard to collect data,” says Michael Hunt, CIHI’s director of pharmaceuticals and health workforce information services. “At the time, the whole culture around adverse event reporting was one of shame and blame, so anonymity was important.”

Annual reports also wouldn’t do. People wanted to see incident data as soon as it was submitted, and an open-access environment to share and learn from each other by communicating in an anonymous way.

Breaking the mold

“From the beginning, this was a very different type of database,” says Sajan. “It was new ground and created a lot of challenges and discussion at CIHI. It meant if you ran a query right now and two hours from now, you’d have different data. That blew people away.”

By 2008, a data set had been established and vetted, and 18 hospitals across Canada were recruited for a five-month pilot project.

“It was a leap of faith,” he admits. “You don’t know until you go out and try it if it’s going to work and be useful. Demands on the time of health care professionals are huge—and this was one more thing to do, so if they didn’t see value, they wouldn’t buy in.” 

Successful pilot

The plan was to pull the system down post-pilot and refine it. But in the end, every hospital asked the team to keep NSIR running—which it did.

The system’s official launch was in April 2010, and today more than 200 facilities have contributed 10,000 records. 

Despite NSIR’s voluntary nature, last summer Ontario’s health ministry mandated its 150 hospitals to report critical incidents to the system.

In Manitoba and Saskatchewan, health regions have embraced it. Testing will soon start on submissions from British Columbia, where a province-wide system captures incidents from all hospitals and long-term care facilities. 

Culture shift 

As a culture of patient safety has taken hold, people are looking more closely at the way things have always been done. Something as simple as improperly stocked shelves are now seen as a near miss, as staff could easily grab the wrong medication in an emergency. For years, hospital pharmacists have fixed prescriptions from doctors but now realize that’s also a near miss.

“We’re seeing more activity as people re-examine their roles,” Sajan says. “The fact that a facility has 800 incident reports this year up from 500 last year doesn’t mean the organization is any less safe or the quality of care is worse. In fact, it probably means the opposite. There’s probably a better culture of safety and staff are more engaged in identifying issues and generating solutions.”

Analytically, NSIR’s also very different.

“You can’t do rates or indicators. I can’t tell you how many incidents occurred in Canada last year, only how many were reported, and that’s very different than any other CIHI database,” he says.

Working with partners

In the growing movement to improve patient and medication safety, this system doesn’t stand alone. The Institute for Safe Medication Practices Canada analyzes medication incident data from multiple sources, including NSIR, to issue alerts and recommendations.

Health Canada will use NSIR data for post-market surveillance and to understand which drugs are involved in product confusion due to similar names or labels.

And the Canadian Patient Safety Institute will use strategies and bulletins developed from NSIR data to support national advocacy.

Still growing

Although NSIR was built for hospitals, some regions have started using it for long-term care. At their request, we did a pilot, determined it worked well and are now recruiting long-term care facilities to the system.

A stakeholder survey will guide which incidents we focus on next. Possibilities include surgical errors, medical devices, radiation oncology and patient falls. 

“NSIR has become more than we ever envisioned,” Sajan says. “In the beginning, we had our vision, but we didn’t know if we’d get it off the ground or how it would be received. To see this value at the local level and beyond is very rewarding.”