One of the best parts of being a HIStalk sidekick is hearing from readers. I’ve got a handful of folks who correspond regularly, sometimes with comments on recent pieces, but other times giving me a heads-up on happenings that I otherwise would have missed. I’ve been way behind on my email due to the amount of work I’ve been doing for one of my new projects and was glad to see something from one of my regular correspondents as I reached deep into the abyss that is my inbox.
Most of us in clinical informatics tend to subspecialize, depending on where we work and how many other clinical informatics professionals are on a project. When I was starting out as a relatively new CMIO, I was a generalist as far as having to work with a lot of different clinical specialties, but was relatively specialized in that 90% of my work was in the ambulatory space. I only had to know about inpatient as much as was needed to address interoperability and the flow of data from health information exchanges and clinical repositories.
From there, I moved to the health system side and had to know a lot more about hospital-based medicine, but still worked with a broad swath of specialties. I became an expert in things like Meaningful Use and clinical quality reporting more out of necessity than anything else.
As a consultant, I run into all kinds of different informatics situations and have to think on my feet. Most of the time I’m fairly well-versed in the topics that get thrown at me, but occasionally I run into something I know very little about. This particular email illustrated one of those situations and was a good lesson on how essential change control really is, along with understanding the downstream impacts of systems changes. My reader was happy to provide some background to get me up to speed.
Over the last 15 years, the US Food and Drug Administration (FDA) has been implementing Risk Evaluation and Mitigation Strategies (REMS) for various medications with serious safety concerns. The goal is to reinforce appropriate use and to help reduce the risks of those medications. One of the most well-known REMS is the program for Accutane, which can cause serious fetal anomalies. As a precursor to REMS, one antipsychotic medication, clozapine, has had FDA-imposed monitoring requirements since its approval in the late 1980s. Clozapine is an antipsychotic drug that can sometimes cause low white blood cell counts, leading to a patient being unable to fight infections. In rare instances, those cell counts can get dangerously low. These effects were seen in the initial clinical trials and frequent laboratory testing was needed before patients could pick up their prescriptions from the pharmacy.
Initially, this was managed using a paper process to submit data to the registry, but for some time, the FDA has had a website where prescribers could enter laboratory results and pharmacists could query whether patients were current. My reader states it worked pretty well, including notifications to providers when a patient was late in having a lab result entered.
However, this changed during the initial months of the COVID-19 pandemic. They note, “FDA left the monitoring up to clinical judgment as patients who were stable on the drug for years really didn’t need monthly labs. But before the pandemic and currently, the rules have been quite clear – no lab test recorded, no dispensing of drug.” This made sense in the context of an emerging healthcare crisis when there may have been barriers to patients obtaining blood work, since having people miss medication doses aren’t good for patients, particularly when missed doses could cause relapse of a serious mental health condition. Additionally, when patients are off this particular medication more than 48 hours, they have to be brought slowly back up to their steady-state dose, which creates a window of suboptimal treatment.
Fast forward to the present, where FDA had an issue with the REMS website vendor that resulted in vendor and process changes. It wasn’t clear to the reader which process changes were supposed to be beneficial as opposed to which ones were caused by limitations of the new website contractor. Regardless, the transition has been described as “an unmitigated disaster.” They note that “the new process is hard to understand, even after taking the mandatory training to register for the new website.” There are PDF forms for submission as well as an electronic process, but the new process is more cumbersome with additional fields, poor layout, and suboptimal usability. Additionally, physicians had to re-enter the results for the most recent blood draw in the new system even though they were in the old system.
To cap it off, the website locked out users even though they had pre-enrolled for the new site, and the help line was overwhelmed, leaving many clinicians, pharmacists, and patients worried that patients wouldn’t be able to get their clozapine. Ultimately, following urgent meetings with stakeholders, the FDA temporarily suspended the documentation requirements. In FDA parlance, “Tthe FDA does not intend to object if pharmacists dispense clozapine without a REMS dispense authorization (RDA).” My reader closes out with this thought — even though the FDA has been focused on pandemic-related matters, they could have handled this transition better.
Putting on my “after-action” reporter hat, it sounds like some key steps were missed, things like stakeholder alignment on business requirements, clinical usability review and sign-off on development requirements, user acceptance testing, go-live support planning, and a contingency plan for reversion or emergent intervention if things were not going well. These are all things that many of us deal with on a daily basis and it’s always baffling how these big projects miss the mark. (Case in point: the VA and Cerner, but that’s a much longer topic and I don’t have enough wine in the house to start tackling that one.)
I hope FDA is able to work swiftly with its vendors to get this sorted out so that patient care can take precedence, and that the learnings from this one will allow them to do better in the future. It’s a good reminder for all of us that work with systems that directly impact patients – we need to be vigilant and make sure that corners aren’t cut so that patient’s aren’t harmed.
What’s the most egregious example of poor change management that you’ve seen in your healthcare IT career? Leave a comment or email me.
Email Dr. Jayne.