Tuesday, January 29, 2013

In Defense of Copy-Forward!

The wonderful folks at HISTalk posted my thoughts "In Defense of Copy Forward" this week (full text below), and as usual - I've had additional thoughts on it... especially when one of my CMIO colleagues said that their auditing folks were actually asking him to look into plagiarism software!   Here was my response to that, as well as some ideas on how we might address the ugly side of Copy Forward (especially on the inpatient side):

Folks - our role as CMIOs is often to serve as the bridge between real-world clinicians and pie-in-the-sky (or at least non-clinically oriented) legal/admin/executives/IT/politicians, etc…   And one of our chief responsibilities is thus to bring everyone back to common sense when hysteria starts to set in.

So please, everyone take a breath - and repeat, "If I am asked to review plagiarism software for my organization, I will tell them they are off their freakin' rocker"… and make them write it down 100 times.   Or maybe I will make a deal, if we use it on medical records, then we can also use it on all their legal documents, managed care contracts, annual reports, etc... again, let's just use common sense!   We are supposed to be using standardized format and structure… so it is expected that notes should be 60-90% similar from visit to visit, or day to day in the hospital.   On the other hand, I know it can get bad - especially on the inpatient side, especially in an AMC where residents, students, fellows and attending are all writing notes!

So what can we do?  Telling docs to not use a key functionality doesn't make sense and is very much the "bad apple" approach of punishing everyone because a few abuse the system.  We need to think about big picture innovations we can do to improve the system for everyone.  I think there are two core issues we need to figure out:
(1) Multiple authors:  For this issue, I'd suggest rethinking how notes are created, and consider a multi-contributed note… similar to a Wiki, but would need to meet the legal standards.   I believe some EMR vendors are exploring the concept of a multi-contributed note, and I do think there is some balance here in making it both easy to use and higher quality than what we currently do… which is often like a mid-1990s version of MS Word.
(2) Poorly trained providers:  I'd put this issue on all of us (GME, Informatics, Clinicians)… I think we have not done nearly as good a job as we should in understanding how to document and then explaining that to those we teach.  And we certainly have not made them feel very responsible.  I think one way to "monitor/measure" this would be to have random chart audits looking for these type of issues, and present them in an "Morbidity & Mortality" style format that will make providers take documentation a bit more seriously… hmm, I actually like that idea!   I hope someone does this and will let me know what happens!

Full text of the original blog:
I’m part of the Association of Medical Directors of Clinical Information Systems (AMDIS), a group of 2,000+ physicians who are the experts in implementing and using EMRs. We have a pretty lively listserv discussion board, and I enjoy seeing what my colleagues are thinking, as well as posting my own thoughts. I especially enjoy posting when I feel like certain studies or comments by non-clinical researchers, administrators, or politicians make us start to question common sense.
One of my favorite topics recently came up — the fear and horror associated with actually reusing some of a previous note. This usually falls into the concept of "Copy-Forward" (when you copy forward the whole note and then edit for today’s visit), or "Copy-Paste" (when you select certain parts of a past note and just copy that part of it. I posted my reply and thought I’d share and expand a bit.
So as not to bury the lead, I think Copy-Forward of a note is a great tool and supports both efficiency and quality, when used appropriately. Turning it off is a classic throwing the baby out with the bathwater analogy. To clarify my biases, my thoughts and ideas are mainly from the perspective of an outpatient physician using Copy-Forward over the past decade, but much of this certainly can be applied to the inpatient world in various ways.
Also, the use of Copy-Paste has some similarities to Copy-Forward, but I agree Copy-Paste is not nearly as efficient and poses more quality issues since it does not have the automatic updating features you might see with Copy-Forward. Here are the points I would suggest we consider.
First, I am sick of these reports which say that things like, "We used plagiarism software to show that 60-80 percent of a doctor’s note is the same as their last one." Um, of course! Since when did progress notes become creative writing endeavors about coming up with different ways to document diabetes, hypertension, and obesity in the same patient visit after visit?
The creative parts of doctoring should involve being "House": figuring out the diagnosis, figuring out the best treatment plan, and artfully explaining it all to the patient. It should not be writing Edgar Allen Poe-like short stories to amuse our auditors or confuse our colleagues. Although, it could be fun, hmmm… what if I described a diabetic’s problems with hypoglycemia in Poe’s style: "Arousing from the most profound of slumbers (due to a glucose of 45), the patient states he feels as if he was in a gossamer web of some dream. Yet in a second afterward, so frail may that web have been, he claims to not remember that which he was dreaming."
Second, there are obvious efficiency benefits to Copy-Forward, but there are very real quality benefits as well. The most obvious is that this type of workflow makes it less likely that important diagnoses will be missed or forgotten over time. Additionally, many systems update certain pieces of data during the Copy-Forward process, so that you can see the most recent results (discussed more below). Obviously incorrect information can be duplicated, especially when a note is being authored by multiple providers over time, but this is where good training and leadership are needed to ensure every provider feels fully responsible for everything in their notes.
Third, getting rid of Copy-Forward or even Copy-Paste is certainly overkill, but we do need to use some common sense in designing technology, workflows, and processes that make it easy to do the right thing when documenting. In the ideal system, much of the critical data would either be updated automatically (e.g. the most recent lab would appear when a note is copied forward), or the system would date entries so it is clear what was done in the past versus today. To clarify, let me break down how an ideal progress note might look like when Copy-Forward is used:
Allergies, Meds, Problems 
These update automatically, which is great, and means the note has the most recent data. I would hope all EMRs have this functionality already.
Past Histories (Social, Surgical, Family) 
These copy forward and allow for easy editing in the note. Ideally, they could be managed in a widget external to the note and have them update from those profiles as well.
Physical Exam 
Want to ideally be able to view old physical exams, and even reuse them when desired (except for vitals). In my current system, the full exam (sans vitals) does copy forward. So I usually just delete it and drop in a new macro and edit that. However, some patients have findings I want to compare from last time (e.g. size of a rash), or consistent findings (e.g. murmur) which I want to be reminded about
For labs (e.g. CBC, chem, chol profile) and certain studies (e.g. mammogram results, last ECG), we use macros which "auto-updatem" so when a note is copied forward, they update automatically to the most recent dates and values.
As some have heard me detail before, I use a form of "problem-oriented charting" in which I type out the history, impression, and plan for a diagnosis (e.g. diabetes) or system/problem area (e.g. "GI issues") all on one line. I also use a macro which includes the date of the entry and my initials.
  • Example for a diabetic patient. "01/19/13(LLB): Stable on Metformin 500bid, CS 100-120s before meals, no med side effects or other complaints. Impr: Stable DM, PLAN: CPM, labs, rtc 4 mos". No flourish is needed. The result is that when copied forward I can see the last time I addressed the DM and if I made any changes. In the same "area" for the problem, I would also have a list of relevant meds, labs, and testing results (e.g. ECGs and ECHOs for hypertension). This way I can see everything I need about a problem all in one place – which means I can make quicker and more accurate decisions.
  • Summarizing old entries over time. I will either retain the old entry, or can summarize over time (e.g. I might take four entries from 2012 and summarize into one line such as, "2012: Dx with DM 4/12, added Metformin 500qd, 6/12 incr to 500 bid and did well").
  • Multiple issues. Since I often address multiple issues in a given visit, I created a line which reads, "Problems below not addressed this visit" so that I can clearly demarcate what I did and did not address on a certain day. I think this method is extremely efficient and higher quality than the method of trying to document all the HPI about multiple issues at the top of a note, and then separating out the Impr/Plan at the bottom.
  • What is a SOAP note? Larry Weed, MD devised the concept of problem-oriented charting 50 years ago, but I think it’s fair to say we have over-complicated it over time. The SOAP note is supposed to be based around a problem. In other words, each problem should have a documentation area for Subjective, Objective, Assessment and Plan. Instead, we create one large SOAP note where we break away all the Subjectives into their own paragraph ("HPI"), thereby distancing your thinking about the complaint and what we are going to do about it. I hope we will soon see more EMRs going "back to the future" by embracing the true problem oriented charting philosophy.
Fourth, the outpatient world is different from inpatient, but there are similarities. I understand that inpatient notes can be more difficult to manage due to quickly changing problems, and especially multiple authors. Personally, I hope we put some more thought into the concept of an "Inpatient Wiki," a single type of inpatient note that can automatically pull in the relevant information for each specialty (e.g. different for medicine, OB, and various types of surgery). Then each author could see what they need to see – it would pull in the labs, tests, consult suggestions, or a nursing note – why make the doctor repeat this themselves every time?
The care provider would then be prompted to write what they are supposed to add, and the note would be a living document which flexes to the individual, but can be time-stamped for medico-legal purposes as well. It could have clear sections (similar to above), as well as an organ or system based areas (e.g. Cardiology issues, GI Issues, Neuro Issues, F/E/N issues) for documenting the SOAP note .
In summary, I would go as far as to say that we need to change our paradigm to "The Note is the Chart." The chart should no longer be a collection of distinct and incomplete notes, but rather the last note can really be the complete chart which contains everything a provider needs. If we do this, then we can reframe our expected workflow from, "You need to read every note ever written to understand the full patient" to, "You just need to read the last note".
The result: when a patient goes to the ER or sees another doc, those providers will find that the most recent note in the system will have all the info they need, so they won’t need to try and dig through 48 notes over 10 years (and let’s face it, they never do that anyway). Granted, the paper record allowed for a much easier way to flip thru past notes, but sooner or later we have to acknowledge that computerized systems have different attributes than paper. We can either keep trying to force the computer to act like paper, which never works out well, or we can start embracing the differences and truly take advantage of them.

Sunday, January 06, 2013

We Don’t Have a Shortage of PCPs, We Have a Shortage of Using Them Efficiently

I've been asked to serve as the "Innovator-at-Large" (aka Editor-at-Large) for the magazine "Clinical Innovation and Technology"... which I was happy to accept as it's the perfect intersection of my worlds!.   In my first post, I've expanded on a phrase I've been using for several years - that "We Don’t Have a Shortage of PCPs, We Have a Shortage of Using Them Efficiently".   I go on to describe the future of healthcare in a world where innovation and IT are being used to their potential to make life easier for physicians and better for patients (of course assuming our reimbursement system equally evolves).  I hope it inspires you!

We Don’t Have a Shortage of PCPs, We Have a Shortage of Using Them Efficiently

Every few months another study warns of a severe shortage of primary care physicians (PCPs) in the future. A recent report published in the Annals of Family Medicine explained how we will require 52,000 more PCPs by 2025 due to population growth, aging demographics and insurance expansion (Reference: 1. Ann Fam Med  2012;10(6):503-509).

Fortunately, both clinical IT and innovation will deeply change medicine over the next decade, resulting in a new paradigm with the potential to improve both efficiency and quality of care. In this paradigm, software will be able to automate or delegate much of the routine care usually provided by physicians. If automated systems and empowered staff members manage stable patients according to evidence-based protocols, physicians can focus on more complex patients who truly require their attention. Individual physicians will actually see fewer patients, but oversee a team who will care for more patients. Thus, we won’t need more physicians; we will just need a better system to help most appropriately leverage physicians, staff and IT.

A typical physician’s office in 2025 might look something like this: Dr. Blake Willoca arrives around 9a.m. and sits in front of a bank of computers and video screens. Dashboards provide real-time analysis of the status of his panel of 5,000 patients. Patients in the Green Zone will be managed mainly by computerized systems which check on patients virtually to provide positive feedback and ensure they stay on track. Meanwhile, patients in the Yellow Zone will be visited by the physician’s care team at home or work, or perhaps have a virtual conference with the physician to answer their questions. Finally, those patients in the Red Zone will be seen in the office or home for longer sessions with the physician and his or her care team to help determine what is going on and how to get it under control. Today, Dr. Willoca will spend an hour with each of these four Red Zone patients in his office, he will do five-minute video conferences with staff members taking care of 20 Yellow Zone patients, and he will spend some time in a virtual reality game teaching med students about how this new system works. As Dr. Willoca leaves his office at 5p.m., he knows he’s helped the patients who most needed it today in a relaxed and livable manner, and he knows that his IT tools and care teams will continue to monitor and help manage his patients 24 hours a day.   

This might all seem like a PCP’s dream, but we need to recognize and accept that we are the generation who will make this happen. There is much to do in healthcare, and there could not be two greater tools to use than clinical innovation and IT.

Online at http://www.clinical-innovation.com/topics/practice-management/we-don%E2%80%99t-have-shortage-pcps-we-have-shortage-using-them-efficiently

As questions come up on this article and topic, I'll make sure to post answers here.

* Barriers: Someone asked why we don't see more of this type of attitude from doctors?  My answer: I think there are two main barriers we need to overcome to increase the spread of this type of "team-based CDS" which automates and delegates clinical work:  First, we need to continue to better align incentives (i.e. Value-based vs. Volume-based reimbursement and legal systems)... How can we expect doctors to delegate work if they are not protected financially or legally from doing just that?  Second, we need to make CDS easy and intuitive to use and ideally integrate them fully and elegantly into our EMR systems (see examples below of companies working on these types of tools).

* Speed: Someone asked why this can't happen sooner (i.e. why did I say 2025 instead of 2015).  It's a good point, as I think it is technically feasible today.  I used 2025 because (1) It was the year used in the article I initially quoted about MD deficits, and (2) I did want to describe a future world where this vision of team-based care and HIT would be completely common and routine, not simply possible.  I think there are financial and legal issues which will slow it down, but I also think that we will be seeing more and more of this happening in the near term as well - just not as widespread and pervasive as we'd like for another 10 years or so.

* Risk Stratification: Someone asked how many patients would fall into the Green/Yellow/Red zones.  Studies have shown that 1% of the population accounts for 20 - 30% of the cost, and 5% account for 50%, whereas the healthiest 50% account for just 3% of the total cost.   So I'd suggest the "Red Zone" is about 5% (e.g. 250 patients in a panel size of 5000), the Yellow Zone would be around 20% (1000 in a panel size of 5000), and then the other 75% in the green zone.  Using another way to measure it, today's typical panel size is 2500, which requires a PCP to see about 25 of these patients in a given day.  If the panel size were 5000, the old system would require them to see 50 patients a day.  In the "new" system, I'd suggest they will need to see about 5 "Red Zone" patients a day in the office while interacting with another 20 "Yellow Zone" patients (or answer questions for their staff members) - which could take anywhere between 1-5 minutes.   The reason that this number stays relatively high is that the Red and Yellow patients do need to be actively managed on an ongoing basis - some will eventually move into Green territory, but others simply have too many interacting or unique problems and medications which are beyond the scope of even advanced protocols and is where the cognitive skills of physicians will shine.  Of course, when artificial intelligence gets good enough to figure all these things and how to communicate it all to patients - then we may see even more automation in healthcare... but if/when computers have gotten that good - we will likely see automation in every other professional career as well - from lawyers and judges, to politicians and marketers, to stockbrokers and Venture Capitalists.  In other words, while I agree with Vinod Khosla's assessment that we will see HIT further automating healthcare... I don't think it will "replace 80% of doctors" - but it will allow us to effectively leverage the current amounts of physicians.

Companies Making "Physician Efficiency Apps" (or "Doctor Happiness Tools" as I like to call them)
* healthfinch: A cloud-based decision support system which integrates with EMRs to automate and delegate repeatable work away from physicians and towards their staff in a safe and consistent manner (e.g. Medication Refills).  I founded this company in 2011 with two very smart HIT experts focused on human-centered design, and have mentioned in some past blogs about "Saving Primary Care with Team-based Delegation Software" and another about "EMR Extender Tools creating Doctor Happiness".  The first product, RefillWizard, which integrates with some of the main outpatient EMRs to help decrease the amount of time doctors must spend approving medication renewal requests, saving them up to 30 minutes daily. If we apply that 30 minutes of savings to the 400,000 primary care physicians in the US, we can effectively "create" 25,000 new physicians—half the expected shortfall in physicians caused by population growth, aging demographics and insurance expansion!  Now just create a few more of these and we save the healthcare system!
* healthloop: Automates the "follow-up" process to check on patients after their in-person visits.  Founded by Dr. Jordan Shlain, another of the rare but growing breed of working PCPs who understands how HIT can help make life easier for docs and patients and is building tools to fulfill that vision.

Other Relevant Articles
Primary Care Physician Shortages Could Be Eliminated Through Use Of Teams, Nonphysicians, And Electronic Communication, Health Affairs, Jan, 2013 (vol. 32, no 1): 11-19.   Says that there will not be a doctor shortage as long as we optimally utilize team-based care and HIT.
Estimating a reasonable patient panel size for primary care physicians with team-based task delegation.  Altschuler J, Margolius D, Bodenheimer T, Grumbach K. Ann Fam Med. 2012 Sep-Oct;10(5): 396-400. doi: 10.1370/afm.1400.  Estimates how much care can be delegated in a team-based model, and thus what an optimal panel size could be to do perfect care.
* Project Doc Shortage is Real, Experts Say.  Modern Healthcare, Jan, 2012.   Discusses that while team-based care and HIT will improve efficiency, we will still have some need for more PCPs - especially in underserved areas.
* Doctor Shortage Getting Worse.  A CNBC article (Mar 13, 2013) where they use the usual claims (again, based on the current model of care) and I appear to be the "poster doc" for the concept of using IT to improve efficiency and save time.  My section: And one expert says it's not so much a scarcity of physicians but of using them in the right way.  "We don't need more physicians, but rather better "team-based workflow tools" to ensure that everyone on the team can work to the highest level of their ability in a safe and efficient manner every day," said Dr. Lyle Berkowitz, Associate Chief Medical Officer of Innovation for Northwestern Memorial Hospital.  "That means using information technology and freeing physicians to spend their time on more complex patients," Berkowitz added.