Digitized Medicine

This case study is the first of six cases used to illustrate aspects of EMR use that pose risks to patient safety.   The risks discussed in this case pertain to EMR hardware and software failures.  As with most EMR-related problems, the majority of patient safety issues associated with hardware/software failures can be traced back to controllable factors.  The safe and meaningful use of EMRs is facilitated by awareness of these risks and controllable factors.

Case Study:  At 0900 physicians in hospital-based ambulatory offices were suddenly unable to access their web-based EMR. The hospital’s IT Service Desk reported a wide outage of internet access due a “hardware failure” and the estimated time of repair was “unknown”.  The offices activated their paper-based downtime plans, but many patient safety issues were encountered.  For example, physicians were solely dependent on the patient’s recollection of allergies because they had no access to allergies recorded in the EMR.  A process was developed “on-the-fly” to use a laptop with mobile broadband connectivity to access the EMR, but before these were ready the original problem was resolved at 1130.  It was discovered that an IT employee installed a planned “Windows update” to the hospital’s networking hardware that morning with unexpected results.  After the problem was identified the update was “backed out” and the problem resolved.

Key Points:

• Hardware and software technology problems will occur, but most are preventable and can be minimized with proactive IT management

Access to secure and reliable patient data in an EMR is dependent on hardware and software that is properly sized, configured, managed and functioning.  Power outages, running out of data storage space, network outages, software conflicts, malicious intrusions and computer device failures are examples of technical problems that threaten safe use of EMRs.  Most technical problems, however, are preventable through proactive IT management.   For example, an old computer that burns out could have been prevented by proactively replacing old technology equipment at specified intervals.  Reputable IT experts know the “best practices” for proactive IT management (such as ITIL and COBIT).  In this case basic change management practices were not followed.  Physicians should make sure that their EMR hardware and software, whether physically located in their office or located elsewhere, are proactively managed by reputable IT experts.

• Physicians are responsible for developing “downtime plans” that describe how the office will safely care for patients when the EMR is unavailable

HIPAA requires patient data to be “backed up” and physicians need to be able to restore their EMR in the event of a complete hardware or software failure such as a flood, tornado or fire that destroys computer equipment and software.  “Disaster planning” addresses these needs and primarily requires the EMR vendor’s technical expertise and advice.  “Downtime plans”, on the other hand, primarily need the physician’s expertise to define what needs to happen in order to safely care for patients (what I call clinical continuity) and maintain business operations (commonly referred to as business continuity) when the EMR is unavailable.  Downtime planning should identify the critical patient data that is needed when the EMR is down and have written procedures on how that data will be accessed.  The plans should also describe procedures for the use of downtime paper forms for patient care tasks, medical record documentation and practice operation tasks (appointments, claims and billing) as well as what to do with them after the downtime.

• “Mock downtime drills” are an effective way to determine whether an office is prepared for an EMR downtime

In this case the physicians deserve applause for developing written downtime plans.  However, if they had previously simulated a downtime, then they would have been prepared with back-up laptops ready and available.  Although mock drills in themselves create risks, they are much more controllable than the risks inherent to unplanned EMR downtimes.

I wonder how many physicians who used social media for the first time yesterday were aware of the risk they were taking?   Did they realize that unprofessional behavior involving social media could actually threaten their license to practice medicine?   These thoughts motivated me to blog about the risks of social media that are unique to physicians.   My intent is not to cause Henny Penny alarmism and avoidance of social media.   After all, I obviously engage in social media for leisure as a practicing physician myself and enjoy it very much.   Instead, I want my colleagues to feel confident using social media, but with full awareness of the inherent risks, all of which are manageable including many that are avoidable.   

I'm just sayin'...

Social Networking Risks

Social networking sites will often publically display personal identifying information unless one actively omits them from public view including home address, e-mail addresses, home phone number, mobile phone number, race, religious preferences, sexual orientation and political views.   Also, when you associate with someone on a social networking site you provide them access to personal content that is not available to others on your public page.   The personal content could include personal identifying information, pictures, personal messages and even messages posted to your page from other people.  

A recent survey (JAMA. 2009;302(12):1309-1315) revealed that a significant number of medical students have personal Facebook accounts with personal identifying information available to the public.  In addition to this privacy exposure, some had also posted unprofessional content on Facebook such as:

• Violations of patient confidentiality
• Pictures of illicit drug use or unethical behaviors
• Use of profanity
• Use of sexually-suggestive language
• Criticisms of other people or institutions
• Satirical or sarcastic material

Posting unprofessional content on a social media website could be judged by a credentialing body, such as the Texas Medical Board, as a reason for disciplinary action that could place the physician’s license to practice medicine at risk.  

Social Networking Risk Management

“Unprofessional content” can be defined to be any content that shows, implies or suggests that a physician has failed to adhere to professional behavior.   Content posted on a social media site that explicitly shows a physician engaged in illegal or unethical behavior is obviously unprofessional.   But how can physicians  distinguish the border that separates "acceptable" content from "unprofessional" content?

Consider the core curriculum on “Professionalism” that the Accreditation Council for Graduate Medical Education (ACGME) uses for physicians-in-training.   This curriculum includes training on professional behavior in areas such as:

1. Compassion, integrity, and respect for others
2. Responsiveness to patient needs that supersedes self-interest
3. Respect for patient privacy and autonomy
4. Accountability to patients, society and the profession
5. Sensitivity and responsiveness to a diverse patient population, including but not limited to diversity in gender, age, culture, race, religion, disabilities, and sexual orientation

The line that distinguishes sensitivity from insensitivity to human diversities is perhaps the most difficult one for physicians to discern when communicating on the Internet.   Content that includes insensitive comments on gender or other human diversities falls into the category of unprofessional behavior.   Unfortunately, the diverseness of individuals and cultures can sometimes make it very difficult to anticipate a perception of insensitivity.   The best advice is to avoid comments that have even a remote possibility of offending someone.

Also, physicians who understand and adhere to patient confidentiality in their daily practice will be vulnerable to breaches of privacy when using social media if they fail to understand the lack of anonymity when using social media sites and the permanence of any content they post.   In spite of using “alias” names, no one is really anonymous on the Internet.   There are many examples of people (including doctors) who presumed anonymity but ended up with serious legal difficulties.   The permanence of posted content is due to the nature of social media.   Even if one deletes their entry, it is possible that other social media sites have already linked to that content and/or reposted it elsewhere. 

The bottom line is that physicians should realize that whatever they write on the Internet not only is permanent, but is also traceable to their actual identity.   Physicians are wise to gain more insights on social media risk management by discussing it with their personal lawyers and seeking information from their professional organizations and journals (such as these articles from the American Medical Association and New England Journal of Medicine).  

Raising awareness of the risks that social media present to physicians is prudent, not alarmist.  I want other physicians to safely enjoy social media as I do.    BTW, I expect you to now read my Terms of Use, Disclaimers, Warranties and Waivers.   And agree to hold this Author harmless.

I'm just sayin'.

In a previous blog the methods used to perform a work flow analysis in a physician's office as a part of the EMR selection process were described.  The  next steps are to:

Document Future Desired Work Flows

An analysis of current work flow identifies bottlenecks in the physician's practice.  Changes in work flow may alleviate those bottlenecks.  Some of the identified changes may not be dependent on having an EMR and could be made immediately.  The other work flow changes that are dependent on using an EMR and will later become part of the EMR implementation.   

It is not readily apparent to many physicians how an EMR could improve work flow in their office.  There may also be unrealistic expectations about how an EMR could improve things.  Therefore, it is helpful to first gain knowledge about the "best uses" of ambulatory EMRs as experienced by other physicians before identifying desired future work flows.  To gain insights on the realistic and best uses of EMRs physicians may be wise to engage a knowledgeable IT consultant.  Alternatively, physicians may use other resources  to gain insights such as:

• DOQ-IT's Operational Redesign Workbook
• AHRQ Health IT Toolbox
• CCHIT
• Networking with colleagues
• Physician professional organizations

The knowledge gained on EMR "best uses" and the identified work flow  bottlenecks can now be analyzed together to determine at a high level what future work flows are desired for the practice. 

Prioritize this list of work flow desires

The physician practice may now discuss these desired work flow changes and prioritize them.  For the purposes of discussion let's suppose the practice has identified a top ten list of desired work flow changes that they want make when they implement an EMR.

Develop a prioritized list of EMR functionalities

The prioritized top ten list of desired work flow changes will naturally translate into a list of EMR functionalities that are top priority.  For instance, a practice that determines "refilling 80 prescriptions/day" is currently their top bottleneck, then the most important EMR functionality to compare among products is the usability of the EMR's e-prescribing feature.  A different practice may identify their top bottleneck to be getting patients through the check-in processes and determine that they want to reduce time patients spend filling out papers on clipboards by providing online registration forms and using a self-serve kiosk in the waiting room.  They also desire an EMR that can quickly register and check-in patients.  In that practice the most important EMR functionalities will be a robust patient portal, efficient integration with kiosks and the fewest number of necessary screens and "clicks" when registering or checking-in patients at the front desk.

Compare EMR products based on these priorities

EMRs have hundreds of functionalities.  It is not humanly possible to effectively compare hundreds of functionalities between different EMRs.  Comparing EMRs without focusing on specific needs easily leads to frustration and/or confusion.  Using the described list of prioritized EMR needs will improve the comparison process. 

Physicians might consider limiting their comparison to EMRs that have been certified by CCHIT (and soon to those certified for "Meaningful Use" as well).  One can be confident that a CCHIT-certified EMR has all of the functionalities that are described on the CCHIT website.  There are hundreds of such functionalities.   However, CCHIT does not necessarily quantify how well the EMR performs each function.  So, if a physician practice limits their comparison to CCHIT-certified EMRs, they can be assured the EMR can do those hundreds of things an EMR should be able to do and instead spend their time comparing how well the EMRs meet their own, identified "top ten" needs.  

Using an earlier example of an identified "top ten" EMR need, a comparison may find that EMR Product "A" allows prescription refills to be completed in 45 seconds using three screens and 17 clicks, while Product "B" refills prescriptions in 15 seconds (30 seconds faster) on one screen with 6 clicks.  Since the work flow analysis revealed that the practice has 80 refills/day, refilling prescriptions using Product "B"" would take 40 minutes less each day as compared to Product "B".  

In summary, it is helpful for a physician practice searching for an ambulatory EMR to perform a work flow analysis to identify the major bottlenecks in the office, gain insights on how EMRs can improve work flow, use this knowledge to develop a prioritized list of desired future work flows, identify the "top ten" things an EMR needs to do to meet those desires and then compare CCHIT-certified products based on this prioritized list of EMR needs.

EMR Selection Guide provides an outline of other topics on the selection process

EMR Implementation Guide provides an outline of topics on the implementation process

cook children