I spent part of the weekend hanging out with some longstanding healthcare IT friends. There was plenty of catching up to do. It was great to hear about everyone’s newest projects and what they think will be hot in healthcare over the coming years. We all used to work together, but now everyone is scattered to the winds at various health systems, consulting firms, payers, and vendors. A couple of have even retired and I enjoyed hearing about their many adventures and how much they’re enjoying not being part of the madness anymore.
Several of us have had experiences helping or parents or other relatives navigate the healthcare system in recent years, so there were plenty of conversations about the usefulness of various patient portals and how our loved ones are or are not using the information available to them. One hot topic was patients having access to their notes from medical visits. Although most of the folks I was speaking with were eager to see their own documents, probably due to the fact that we’ve spent so much of our careers dealing with EHRs and other clinical IT systems, very few of their parents show an interest in reading their medical notes.
The clinicians in the conversation voiced concerns about patients being able to see their notes and “how bad they really are” due to EHR-related note bloat. There were few concerns about patients reading things they would find worrisome or being offended by discussion of hot button issues such as obesity, smoking, alcohol use, or drug use. Nearly all felt that the benefits of transparence outweighed concerns, although one clinical informatics nurse lamented the fact that many of the clinicians that she supports write notes that can be described as “terrible,” including typos, sentence fragments, and disordered thought processes. I certainly feel for that physician’s patients if they try to access their information.
With that conversation fresh on my mind, I was happy to see a link in my email to a recent article in the Journal of the American Medical Informatics Association that looked at the impact of patient access to notes on clinical documentation. The authors set out to examine whether recent rules that require patient access to clinical notes have had unintended consequences, such as increasing physician documentation burden. They used what they describe as “a national, longitudinal data set consisting of all ambulatory care physicians and advance practice providers using an Epic Systems EHR” to specifically evaluate the length of clinical notes and the time spent documenting in the EHR.
As background, the 21st Century Cures Act language that required that healthcare organizations provide access to notes was implanted on April 5, 2021. The authors used “de-identified clinician-week level EHR audit log metadata extracted by Epic’s Signal software” to look at aggregate data from January 3 to May 29, 2021. This data covered over 340,000 unique clinicians for 21 weeks and excluded inpatient data. Analysis models controlled for the number of visits that clinicians were performing each week as well as other factors that might impact individual productivity. The authors also checked their data by looking at only primary care physicians and also by stratifying to compare physicians against midlevel providers. During the initial post-rule period, there was no statistically significant change in note length or documentation time.
Interestingly, providers spent more than 12 minutes creating each note, which is significant given the fact that a large number of visits delivered in the US are scheduled for approximately 15-minute appointment times. In the discussion, the authors acknowledge that they were looking at the potential for short term impacts and that more work is needed to evaluate whether there are long term impacts on having patients access their notes. Although I like how the authors approached the problem as well as their use of a large and available dataset, I wonder how many of the organizations in that dataset were actually releasing their notes at the time of the study. My own health system didn’t start releasing notes to patients for at least six months after the rule went into effect, although they did release content retroactively, which was a surprise to me as a patient.
Now that I have access to my notes, I look at them often. I have found their content to be largely fictional. My last preventive health visit contained two full pages of screening questions that were never asked during the visit as well as exam elements that weren’t performed. There were also inaccuracies in my responses to questions, along with what I find to be a very annoying disclaimer that the documentation “was created with voice recognition software and may contain errors or omissions.” Seriously? I can’t imagine having a patient see that information in one of my notes, and it definitely makes me think less favorably about the clinician I saw, especially since I wasn’t happy with the visit in real time.
I wonder how many patients are actually looking at notes that are available and whether they’re communicating with clinicians when they find problems. I know I don’t have time to deal with trying to get a correction, so I’m going to just let it be even though those inaccuracies will likely propagate themselves across other entries in the future. Even thinking about trying to correct it is tiresome.
I would love to see research looking at how many health systems are actually releasing their notes, how many have communicated to patients that the notes are available, and how many are encouraging patients to use the information to better their health. I suspect that the results of those kinds of efforts would be rather interesting.
The authors note that they are also unable to quantify non-note work that may have increased due to the availability of patient facing notes, such as increased patient phone calls or portal messages. I know that when I put on my CMIO hat to approve patient-facing documentation, I always try to make any default language as clear as possible to avoid creating confusion. However, I’ve seen plenty of notes where clinicians go crazy with free text and create plenty of confusion that I will never be able to influence.
Who’s ready to look at this data again, and see what it looks like 12 to 18 months into the process? Leave a comment or email me.
Email Dr. Jayne.