The Journal of Family Practice published an original research article this month looking at notes written by medical scribes. Since we have scribes in our practice, this was of definite interest.
The authors used the Physician Documentation Quality Instrument (PDQI-9) to look at the quality of notes written by 18 primary care physicians prior to scribe use as well as after the introduction of scribes. The study controlled for type of visit (diabetes visits and same-day appointments) and allowed a period of adaptation (three to six months) after the introduction of scribes and looked at just over 100 notes for each period.
Although it makes for a relatively small sample size, the authors found that scribed notes were “more up-to-date, thorough, useful, and comprehensible” among the diabetes visits. Interestingly, they did not find a difference in quality on the problem-focused same-day appointments. The notes were found to be similar in total word count.
The scribes used in the review were medical assistants acting as scribes rather than an independent scribe. Care teams trained for the new model by having the physician and scribes attend two training sessions (two hours each) and a half day of observation and evaluation in the clinic.
I have to admit, I wasn’t familiar with the PDQI-9 instrument. The authors admit that while it is a validated tool, “it relies on subjective ratings of note quality by the reviewer.” They attempted to control for this by having two reviewers (an internal medicine resident and an experienced internal medicine physician) independently rate the notes and then discuss. Once they found that there was >70 percent agreement on the reliability of the ratings (about 20 notes), the resident was deemed “reliable” and allowed to evaluate a random sample of notes to form the basis of the review.
The authors noted concerns about over-documentation when using EHR-based templates. Interestingly, they also noted that “both physician and scribed notes were rated to be of average to low quality because none of the mean scores on the nine individual components of the PDQI-9 reached 4.0.” That would lead the reader to believe that there is opportunity for improvement in documentation across the board, whether scribed or not. Considering that the push over the last 20 years has been “documentation for payment” rather than “documentation for clinical value,” I’m not surprised.
They also noted some potential drawbacks to scribe use, such as lack of EHR innovation since physicians are shielded from poor EHR usability by scribes. I’m not sure that I agree with that assertion. We use scribes in our practice and have documented data on how they impact physician productivity. We also know exactly how excessive clicking in the EHR hinders scribes and we haven’t stopped pressing our EHR vendor just because we use scribes.
In my experience, physicians in a private practice model or even in an employed model where they are responsible for covering their own overhead are sensitive to the scribe’s productivity and will continue to push the vendor for improved application performance.
The authors also note that “incorporating scribing into a practice may also improve the physician experience, a possible benefit that we did not measure.” Although we do have scribes in our practice, individual physicians aren’t always guaranteed to have one. Our scribes are deployed to the locations seeing the highest volumes at any given time. They might work at two or three locations in a given day, following the ebb and flow of patients across the city.
Our scribes definitely improve physician satisfaction, so when we’re lucky enough to get one, we try to hold onto them. As the practice has grown, this has led to a need to have centralized management of the scribes, where a team leader looks at the bed boards across the sites, looks at the patient mix, and makes adjustments as needed rather than waiting for physicians to request or release a scribe.
The publication also notes that although all providers used the same EHR, there may be variations in individual provider templates. Our practices has a single set of templates across the organization, so we don’t see that issue. Having a single set seems to help the scribes be more interchangeable given our staffing model. Sure, we have our favorites, but the preferences are likely more about personality rather than speed or accuracy.
I know that when I have a scribe, generally the entire note is done when I walk out of the patient room unless labs or diagnostic imaging is involved. In those cases, the scribe returns to the patient room with me to discuss the results and plan of care.
Even during the most intimate of exams, I’ve not had patients resist the idea of a scribe, especially when the scribe can also serve as chaperone or assist with a procedure to help it go more quickly. That’s definitely an advantage of having dual-trained scribes who can perform other clinical duties. Patients seem appreciative that I’m focusing on them and their needs and am not distracted by the computer.
I may not be the best indicator of that, however, because even when I don’t have a scribe, my ability to focus on the patient is probably better than that of the average physician. Thanks, Mom, for making me to learn to touch type. It’s not only a great skill for patient care, but also allows me to multitask during meetings and make it look like I’m attentively taking notes.
In doing the modeling for primary care physicians, we sometimes find that physicians can “afford” to have a scribe by deploying their existing staff in a more efficient manner. Sometimes that means redistributing work and sometimes it means moving people to different job roles, both of which can be challenging for practices from an interpersonal and political standpoint. As I tell my clients, though, I’m happy to be their bad guy and help them make the change. I’ve even worked with a couple of larger groups to put together a scribe training program and help them get current staff transitioned.
I really like the training model that our practice has – all scribes are personally trained by the physician owner and are only allowed to graduate to other sites with his approval. It ensures consistent quality, but is not likely reproducible in other practice settings. We also use a variety of types of clinical assistants as scribes – medical assistants, paramedics, EMTs, and premedical students. Having this real-world experience has helped me assist my clients in thinking outside the box.
The authors conclude that as use of scribes increase, more research is needed. I definitely agree and look forward to seeing how we work with scribes in the next five years.
What do you think of scribes? Email me.
Email Dr. Jayne.