EHaaaRrrggggh!

Our Chief Medical Information Officer (CMIO) and I recently had a conversation around an upcoming roll-out of a new upgrade on our enterprise-wide Electronic Health Record (EHR).  It was a good discussion, and I think our team has done a great job planning for this major system upgrade. After our discussion though, I started thinking a little more about the EHR in general.  In spite of their well-documented benefits (most notably, decreased costs and improved outcomes), EHRs have also been a huge source of frustration among providers. 


Don't get me wrong - I am thrilled that we are no longer using paper records in hospitals.  I have witnessed firsthand how much of an impact that our EHR has had on reducing medication errors, improving communication among the different providers on the health care team (especially now that physicians and other providers on the team can actually read the hospital charts!), and improving the coordination of care by allowing easy access to past medical records, as well as access to a patient's current medical data from anywhere in the hospital.  However, for all of their benefits, EHRs have also complicated care in ways that were possibly unanticipated.  For example, the cost of implementation of an EHR is incredibly high - while most hospitals can afford the upfront and ongoing investment of resources required to implement and maintain an EHR, the cost for many office-based physicians is still way too high.  In addition, there is always a significant learning curve between different systems at different institutions (in the latter case, there is little in the way of interoperability or transferability even when the different institutions are using the same EHR vendor).


The learning curve associated with a new EHR can be difficult for sure.  Even more concerning is the fact that in many cases, the EHR is not adapted to the normal workflows used in the hospital or office settings, leading to costly inefficiency.  For example, one study showed that office-based physicians spent only 27% of their time on direct patient care and 49% of their time on EHR and desk work.  Another study showed that for every hour that a physician spent on direct patient care, he or she spent nearly 2 hours on the EHR - these same physicians spent an additional 1 to 2 hours on EHR work at night while at home!  As a result, several studies (here's one) have suggested that EHRs lead to lower job satisfaction and increased work-related stress for providers, and many of these studies cite EHRs as the single most important cause of professional burnout.  An article by the late Pulitzer Prize-winning journalist Charles Krauthammer in the Washington Post even suggested that the EHR was driving many physicians to leave the practice of medicine!


With all of this in mind, and in the context of my recent conversation with our CMIO, I found an interesting article in the pile of unread articles on my desk (remember one of my older posts, "Today's word is...Tsundoku"?) this morning entitled, "How Does the Implementation of Enterprise Information Systems Affect a Professional’s Mobility? An Empirical Study".  The authors of the study (from the Mendoza College of Business at the University of Notre Dame) conducted an investigation to determine whether the implementation of an EHR result in increased turnover of physicians (i.e. physicians leaving one institution for another).  They utilized data from both the Florida Agency for Health Care Administration and the Healthcare Information and Management Systems Society (HIMSS) Analytics Database, so the study was focused only on hospitals and physicians in Florida.  That being said, the final dataset included almost 13,000 physicians working at just over 300 hospitals between 2000 to 2010, for a total of just over 144,000 physician-hospital-years worth of data.


There were three key findings reported in the subsequent paper:


1. If the newly implemented EHR provided what the authors' termed non-trivial complementarities (basically, if physicians thought that the EHR produced benefits that made care more efficient or easier), physicians were more likely to stay at the institution.  Conversely, if the newly implemented EHR significantly disrupted normal workflows and routines, physicians were more likely to leave and go to a different institution.  No real surprises there.


2. Younger (which the authors' defined as those physicians practicing for 10 years or less) and older (defined as physicians practicing for 30 years or more) were far more likely to leave when the EHR disrupted workflow compared to mid-career physicians (those practicing for 10-30 years).  In addition, general practitioners were more likely to leave compared to specialists. 


3. The level of competition and number of previous EHR or IT implementations significantly influenced whether or not physicians left.  In other words, if there were other hospitals close by or if there had been a number of previous EHR or IT implementations at the same hospital, physicians were more likely to leave.


So what is the take-home message here?  First, health care leaders need to recognize that the execution of any EHR implementation or upgrade must go well - therefore, plan ahead!  While the implementation team may get a free pass the first time around, subsequent implementations or upgrades that don't go well will likely increase physician turnover.  Second, involving key stakeholders in the EHR implementation is critical.  Here, physicians, as key constituents, should participate during the initial design and build phase of the implementation.  I was just shadowing one of our ambulatory physicians the other day and noticed that she had to go through a number of steps to document her patient's clinic visit.  There are a number of short-cuts (smart phrases, drop-downs, etc) that could have significantly improved the efficiency of her documentation.  Having end-users be a part of the design and build goes a long way towards aligning the EHR with the normal workflows and routines. 


The study's dataset only went through 2010 - I suspect that more health care systems have already made the change to an EHR.  So, I do wonder if physicians are less likely to leave one institution for another, now that both likely have an EHR.  Regardless, I do think that this particular study sheds light on an important topic.  The EHR has clearly improved documentation for revenue capture (here's one of many available studies on the impact of EHRs on revenue capture), now we need to make sure that the EHR works just as well to make documentation better for providers.