Hard lessons learned: VA develops 'look-back' model

Patient notification a complex issue

The Veterans Health Administration has developed best practices in handling large-scale epidemiologic look-back investigations, including finding a way to explain a potential exposure of blood-borne viruses to a large number of people who likely were not impacted by the incident.

The VA developed the processes after a 2009 discovery of incidents in which improper endoscope reprocessing procedures occurred at four VHA facilities over a six-year period, impacting more than 10,000 patients.1

"We had to do look-back investigations in a short period of time," says Gina Oda, MS, CIC, associate director of the office of public health surveillance and research at the VA Palo Alto (CA) Healthcare System. "We learned the hard way because we didn't have all the processes and procedures in place. So we learned the right and wrong ways of doing these kinds of large-scale look-back programs, then we developed a manual and key things to make it go smoothly."

Five key steps

Here are the five main steps the VA developed for a look-back investigation:

1. Identify and notify affected patients.

"The VA now has a large-scale process involving many offices in which we determine in the beginning whether this event involves exposure to patients and places them at risk," Oda says. "Look-back means we go back and look at their medical histories and perhaps bring them back in for additional lab testing to see if they developed some sort of infection from exposure."

The VA determines which pool of patients is involved and then carefully decides how to make the notification to the potentially impacted individuals.

"You have to be very careful because you don't want to notify someone who wasn't involved and scare the person," Oda says. "You have to weigh the risks with benefits and make sure there is enough of a risk that it is worth going back and testing that patient to find out if they are infected."

The VA determined that calling patients on the telephone, while ideal, is not practical in all cases.

"Patients appreciate a personal call, but that can take a long time, so it depends on how serious the problem is," Oda explains. "We don't send out emails; we send out certified letters or make telephone calls."

It's also a good idea to keep track of the people who were contacted, how they were contacted, and whether they came in for testing.

2. Provide services to patients responding to notification.

It's important to standardize the response. The VA set up call centers and trained people to answer veterans' phone calls to inform them further of the exposure event, Oda says.

The call center staff needs to know the basics of what happened and what the health system is doing. It's important the people answering calls can provide reassurance and explain how there is no immediate concern or risk of further infection. They also should provide practical information about when local clinics would be open and what the person needs to do to be tested and learn the results.

"The exposures we've had were not the type of exposures where you knew immediately that something happened," Oda notes.

Usually the risk was minimal, resulting from medical equipment that was improperly reprocessed, but the problem wasn't identified for a long period of time.

However, when the notified patients returned for testing, a number of positive test results were found.

The VA conducted further testing of the viruses to see if there was a link to any potential VA reprocessing exposure and found no connection, Oda says.

"If we had found the virus was the same we could say with high probability that the patient with newly-identified virus actually got an infection from the procedure," she explains. "In none of the cases we tested did the patient have the same virus, so the likelier explanation was the patient got infected some other way."

The problem was that patients were unconvinced that they had contracted HIV or hepatitis from another source, and often they had not been tested previously for the viruses, she notes.

"Unfortunately, we uncovered the underbelly of care, which is that people are not getting tested when they should be tested," Oda says.

"You uncover all these people who are newly positive, and they think they got it from the procedure," she adds. "So we had to deal with quite a few people who were upset and thought they had gotten the virus from the VA, even when we told them it was unlikely."

The VA provided care and follow-up counseling to all of the patients.

3. Follow laboratory testing strategies.

When patients visited clinics in response to notifications, they were tested for HIV or hepatitis C. One lesson the VA learned from this experience was that they needed to provide better instructions to clinics about how they needed to have these blood samples taken.

For instance, local clinicians collected the blood samples, ran the basic screening tests, and if a test came back with positive results they'd call the patient and ask the patient to return for a follow-up blood test, Oda explains.

"It would have been easier if they had saved the blood on all of the patients, so they wouldn't have had to call some patients back for further testing," she says.

The VA learned from this experience and had clinicians collect extra blood samples the next time there was a look-back investigation.

The follow-up blood tests were necessary to determine whether or not patients had obtained their viral infection from the reprocessing procedure, and the VA informed patients upfront of this investigation.

"We informed them that if they were positive we would like to do a test to determine if they got their virus from the procedure, and that was why we saved blood to use for testing," Oda says. "They signed a consent to agree to do that."

4. Disclose test results and provide clinical follow-up.

Clinicians set appointments to discuss the results with patients, whether the results were positive or negative.

"If the results were positive, we set them up with appropriate clinical care," Oda says.

Those testing positive for hepatitis C were referred to a hepatitis C clinic where they received treatment. Patients who tested positive for HIV were sent to the HIV clinic.

If patients refused to go to the clinic where they were initially referred, clinicians found another VA clinic where they could receive care, Oda adds.

5. Conduct an epidemiologic investigation of patients with newly identified infection.

This investigation occurred through the saved blood samples. Once the VA learned that clinics were not conducting the blood sampling as efficiently as desired, the VA developed a testing protocol for them.

"We ended up having them send the samples to us so we could be responsible for additional testing," Oda says. "We're still evolving that process; it gets a bit complicated."

In some cases, the event occurred recently enough that patients would have to return for a second blood test because the first negative test might have been within a window period for the virus, she adds.

The VA's look-back process also resulted in development of a VHA Lookback Program Operations Manual, which still is in rough draft form. When it's completed, the VA will make it available to others, Oda says.


  1. Oda G, Schirmer P, Lucero C, et al. Development of a standardized process for conducting large-scale epidemiologic lookback investigations following improper reprocessing of reusable medical equipment. Abstract 127. Presented at the 38th APIC Annual Educational Conference & International Meeting.June 27-29, 2011, Baltimore, MD.