Free EMR Newsletter Want to receive the latest news on EMR, Meaningful Use, ARRA and Healthcare IT sent straight to your email? Join thousands of healthcare pros who subscribe to EMR and EHR for FREE!

Problematic Medical Bills Drive Consumers To Cut Back On Care

Posted on November 7, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

As we all know, patients and families are taking over responsibility for a steadily greater percentage of their healthcare costs over time. Not surprisingly, this can affect their medical decisions in negative ways. In fact, a new study has documented that their medical bills are confusing or unexpected, a patient may get overwhelmed and simply skip some forms of care entirely.

The study, which was conducted by Hanover Research and sponsored by HealthSparq, surveyed more than 1,000 Americans on their experiences with unexpected medical bills. The results should be unsettling to anyone in outpatient services, especially those in primary care.

Researchers found that more than half (53%) of respondents had received a surprise medical bill over the past 12 months. This included bills that were higher than expected (60%), for services they thought were covered by insurance but weren’t (62%) and from multiple providers when they expected to get just one (42%).

When faced with these frustrating billing situations, patients may drop out of their care routine to some extent. Many skip routine checkups (40%), routine health screenings (39%) or care for injuries (39%).

A substantial number of respondents (40%) conceded that they could’ve avoided such shocks by doing more to better understand their benefits and healthcare processes, in addition to blaming their insurers (45%) or their health providers (42%). Regardless, it appears that a large number didn’t know who was responsible for the problem, which doesn’t bode well for their future health behavior.

Look, everyone knows that offering an accurate estimate of patient financial liabilities could be a nightmare in some situations, particularly if insurance companies don’t play nicely with the billing department. It’s also true that in some cases, patients simply won’t be able to pay the bill regardless of how you present it, a problem you certainly can’t surmise on your own as a medical practice.

That being said, you can take a look at the bills your practice management system produces and get a sense whether they’re decipherable to those who don’t work within the organization. Even if the PM system does a good job of supporting your end of the process, that doesn’t mean it’s turning out bills that patients can use and understand.

Yes, arguably the most important thing a practice management system does is to support your claims process effectively, but seeing to it that patients aren’t overwhelmed by their bills is clearly a big deal too. Particularly under value-based care, you can’t afford to have them holding off on the services that will keep them well.

Stanford Offers 10-Year Vision For EHRs

Posted on October 12, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

Despite many efforts to improve EHRs, few physicians see them as adding value to the practice. Sadly, it’s little surprise given that many vendors don’t worry much about what physicians want, focusing instead on selling features to CIOs.

As a result, they still don’t like their EHRs that much. In fact, a recent survey conducted by Stanford Medicine and the Harris Poll found that 44% of physicians said that the top value of the EHR was to serve as digital storage, which isn’t a ringing endorsement. Just eight percent saw the EHR as having clinical value, with three percent citing disease prevention, 2% clinical decision support and 3% patient engagement as top benefits.

Is it possible to create a new EHR model that physicians love? According to Stanford, we could build out an ideal EHR by the year 2028.

In Stanford’s vision, clinicians and other healthcare professionals simply take care of the patients without having to think about health records. Once examinations are complete, information would flow seamlessly to all parties involved, including payers, hospitals, physicians and the patient.

Meanwhile, it would be possible to populate the EHR with little or no effort. For example, an automated physician’s assistant would “listen” to interactions between the doctor and the patient and analyze what was said. Depending on what is said in the room, along with verbal cues of the clinicians, it would record all relevant information in the physical exam.

What’s more, the automated physician’s assistant would have AI capabilities, allowing it to synthesize medical literature, the patient’s history and relevant histories of other patients available in anonymized, aggregated form.

Having reviewed these factors, the system would then populate different possible diagnoses for the clinician to address. The analysis would take patient characteristics into account, including lifestyle, medication history, and genetic makeup.

In addition to its vision, the survey report offered some short-term recommendations on how medical practices can support physician EHR use. They included:

  • Training physicians well on how to use the EHR when they’re coming on board, as well as when there are incremental changes to the system
  • Involving physicians in the development of clinical workflows that take advantage of EHR capabilities
  • Delivering EHR development projects as quickly as possible once physicians request them
  • Making data analytics abilities available to physicians in a manner that can be used intuitively at the point of care
  • Considering automated solutions to eliminate manual EHR documentation

Technologists, for their part, can take also take immediate steps to support physician EHR use, including:

  • Developing systems and product updates in partnership with physicians
  • Limiting the use of manual EHR documentation by using AI, natural language processing and other emerging technologies
  • Using AI to perform several other functions, including synthesizing and summarizing relevant information in the EHR for each patient encounter and offering current and contextualized information to each member of the patient care team

In addition, to boost the value of EHRs over the long-term, 67% of physicians said making interoperability work was important, followed by improving predictive analytics capabilities (43%), and integrating financial information into the EHR to help patients understand care costs (32%).

Quality Payment Program Tops List Of Regulatory Burdens On Medical Practices

Posted on October 10, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

A new survey by the Medical Group Management Association has found that meeting the demands of the Medicare Quality Payment Program tops the list of regulatory burdens named by respondents in medical practices.

The survey, which collected responses from 426 medical groups, found that their regulatory burdens were climbing, with 86% reporting that such burdens had increased over the past 12 months. A smaller but similar share of respondents (79%) reported that the overall regulatory burden associated with participating in Medicare specifically had increased during the same period.

When asked to name the regulatory requirements they considered to be very or extremely burdensome, 88% named the Quality Payment Program, followed by prior authorization (82%), lack of EHR interoperability (80%), government EHR requirements (77%) and audits/appeals (68%). In contrast, just 49% of respondents saw compliance with HIPAA privacy and security requirements to be a major concern.

Given the challenges it imposes on practices, it’s no wonder that the MGMA respondents struggle with MIPS, with just 9% stating that they were satisfied or very satisfied with the performance feedback the program offers. Two-thirds of respondents told the MGMA that at least in its current form, MIPS doesn’t support their practice’s clinical quality priorities.

Perhaps the most irksome aspects of the MIPS program seemed to be the full-year quality reporting period and scoring methodology. Roughly two-thirds of respondents were dissatisfied or very dissatisfied with these aspects of the program. “The lack of clarity and constant readjusting of the MACRA regulations regarding MIPS/APMs is also frustrating,” one group member said.

In addition, despite ongoing efforts to support patient data exchange, the percent of respondents who rated a lack of EHR interoperability as very or extremely burdensome has climbed over the last 12 months, from 68% last year to 80% in 2018.

Ultimately, this problem could have serious financial consequences for some organizations. “Interoperability will never be achieved at the rate we’re going without bankrupting most private medical practices,” wrote one respondent. “As each of the EHR vendors moves towards their own interpretation of interoperability, they create different versions of their own software that cost all of us more to implement and we can’t afford any more.”

If these issues aren’t addressed, it seems likely Medicare’s drive toward value-based payment will be less successful than its leaders would hope.  Seventy-nine percent of practices responding to the MGMA survey said they didn’t think the move toward value-based payment had been successful to date, and it doesn’t seem likely that this will change if physicians continue to feel overburdened and misunderstood

A Next Step For Personalized Medicine? Vendor Brings Genomics To Ambulatory EHR

Posted on October 8, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

Most physicians have some sense of the value personalized medicine can bring to their practice, but I doubt that many have ready access to the tools they’d need to harness its power.

In an effort to close that gap – and of course, to make its platform irreplaceable – a vendor serving medical practices has struck a deal giving physicians the ability to order genetic tests and leverage them to improve care.

The vendor, DrChrono, offers a suite of electronic systems for physicians, including an EHR which can be customized by bundling in affiliated apps. Its new partner is Genomind, a personalized medicine platform offering genetic testing for psychiatry practices.

Physicians using DrChrono will have access to two Genomind test kits, along with some analytics tools they can use to make use of the testing data.

One of the tests is Mindful DNA Professional, a genetic test used by clinicians to help them guide wellness decisions. The test targets aspects of a patient’s genetic details which could have an impact on overall health, such as variants suggesting that they could have sleep issues or a predisposition to anxiety, depression or impaired cognition.

DrChrono users will also have access to the Genecept Assay, the results of which can guide the treatment of psychiatric conditions. Once test results become available on the Genomind system, doctors can use its gene-drug-environmental interaction tool, the Genomind Drug Interaction Guide, to inform their treatment decisions. With the help of the Guide, clinicians can analyze the patient’s current medication regimen and flag gene-drug interactions.

An interesting side note to all of this is that the final test results from Genomind will be stored in the DrChrono information library for the patient and become part of the patient’s medical record.

Looked at one way, sharing the Genomind test results seems almost like a no-brainer in a world where casual genetic testing (think 23andMe) is becoming the norm. On the other, though, I don’t want to gloss over the fact that using genetic data to search for relatives is one thing and putting it into your personal medical record is quite another. It suggests that of consumer-driven demand for precision treatment is maturing, and that Genomind is on the right side of this trend.

This takes me back to DrChrono, which while not itself reinventing the wheel has struck a smart deal here. Not only has it brought a tool on board which could offer some benefit to physicians, its supporting the collection of information (genetic data) that patients are beginning to want. If DrChrono can give patients their genetic info via a decent portal, the company may find itself to be in demand with patients. Way to stay abreast of the times.

Practice’s EMR Implementation Drove Up Costs For Six Months

Posted on September 28, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

Everyone knows that providers incur EMR-related costs until well after it is implemented. According to a new study, in fact, one medical incurred higher costs for six months after its implementation.

The study, which appeared recently in The Journal of Bone & Joint Surgery, calculated the impact of an EMR implementation on labor costs and productivity at an outpatient orthopedic clinic. The researchers conducting the study used time-driven activity-based costing to estimate EMR-related expenses.

To conduct the study, the research team timed 143 patients prospectively throughout their clinic visit, both before implementation of the hospital system-wide EMR and then again at two months, six months and two years after the implementation.

The researchers found that after the first two months, total labor costs per patient had shot up from $36.88 to $46.04.

One reason for the higher costs was a growth in the amount of time attending surgeons spent per patient, which went up from 9.38 to 10.97 minutes, increasing surgeon cost from $21 to $27.01. In addition, certified medical assistants for spending what time assessing patients, with the time spent almost tripling from 3.42 to 9.1 minutes.

On top of all of this, providers were spending more than twice as much time documenting patient encounters as they had before, up to 7.6 minutes from 3.3 minutes prior to the implementation.

By the six-month mark, however, labor costs per patient had largely returned to their previous levels, settling at $38.75 compared with $36.88 prior to the installation, and expense which remain at the same level when calculated at two years after the EMR implementation.

However, providers were spending even more time documenting encounters than they had before the rolling, with time climbing to 8.43 minutes or roughly 5 minutes more than prior to the introduction of the EMR. Not only that, providers were spending less time interacting with patients, falling to 10.03 as compared with 14.65 minutes in the past.

Sadly, we might have been able to predict this outcome. Clearly, the clinic’s EMR implementation has burdened its providers and further minimized time the providers spend with their patients. This, unfortunately, is more of a rule than an exception.

So why did the ortho practice even bother? It’s hard to say. The study doesn’t say what the practice hoped to accomplish by putting the EMR in place, or whether it met those goals. Given that the system was still in place after two years one would hope that it was providing some form of value.

Truthfully, I’d much rather have learned about what the clinic actually got for its investment than how long it took to get everyone trained up and using it. To be fair, though, this data might have some relevance to the hospital systems that manage a broad spectrum of medical practices, and that’s worth something.

A Vote In Favor Of Using Scribes

Posted on September 26, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

Over the past few years, using scribes to complete medical documentation in EHRs has gotten mixed reviews. Some analyses have found that scribe services were too expensive to justify the investment, while others have concluded that the use of scribes can make a meaningful impact on revenue and improve physician productivity.

This month, a new paper has been published whose results fall into the plus column. The small study, which appears in JAMA Internal Medicine, looked at the use of scribes among 18 primary care physicians.

To conduct the study, researchers looked at physicians at two medical center facilities within an integrated healthcare system, gathering data between July 1, 2016 and June 30, 2017.

The research team assigned PCPs randomly to two groups, one with and the without scribes, for 3-month periods, switching physicians between the with and without groups every three months. At the end of each three-month period, the PCPs filled out a six-question survey which collected their perceptions of documentation burdens and visit interactions.

In addition to capturing PCP perceptions of scribe use, researchers also collected objective data, including time spent on EHR activity. They also surveyed patients of participating PCPs to gather data on the patients’ perceptions of visit quality.

When all was said and done, the research team found that scribed periods were associated with less self-reported after-hours EHR documentation work.

Researchers also found that when they used scribes, PCPs were more likely to report spending more than 75% of the visit interacting with the patient and less than 25% of the visit on the computer. In addition, physicians were more likely to finish their encounter documentation by the end of the next business day during scribed periods.

What’s more, 62.4% of patients said that scribes had a positive effect on the visits, while just 2.4% said they had a negative effect.

The researchers’ take away from all this was that the use of medical scribes could be one strategy for improving physician workflow and primary care visit quality.

As I noted previously, other research has drawn similar conclusions. For example, a study published in 2015 (which included the involvement of scribe provider ScribeAmerica) found that scribe use at the two hospitals was linked to an improved Case Mix Index which ultimately led to gains of about $12,000 per patient. Meanwhile, inpatient physicians were able to cut time spent the chart updates by about 10 minutes per patient on average.

Having been over arguments for and against scribe use, my personal conclusion is that working with them can be a worthwhile investment if doing so is a good fit for the physicians involved, but doesn’t work in all cases.

Ultimately, it seems that there’s too much variation between settings in which scribes could be used to make a single blanket statement about their benefits. I guess we won’t be drawing grand conclusions about scribe pros and cons anytime soon.

Don’t Be The Last Practice To “Get” Digital Health

Posted on September 14, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

Physicians, are you savvy about the digital health technologies your patients use? Do you make it easy for them to interact with you digitally and share the health data they generate? If not, you need to move ahead and get there already. While you may be satisfied with sidestepping the whole subject, patients aren’t, a recent report suggests.

As you probably know already a growing number of patients, most notably millennials, are integrating digital health tools into their everyday lives.

Research from Rock Health, which surveyed about 4,000 consumers, found that the share of respondents using at least one digital tool (such as telemedicine, digital health tracking apps or wearables) hit 87% last year. To get a sense of how impressive this is, bear in mind that just five years ago, only a tiny handful of consumers had given any of these tools a try.

What’s also of note is that some of these consumers were willing to skip insurance and pay out of pocket for digital care. One particularly clear example of this involves live video telemedicine; Sixty-nine percent of consumers who paid out of pocket for such consults said they were “extremely satisfied” with the experience.

Patients who reported having a chronic health condition seemed less likely to use digital tools to track their health metrics. Case in point: When it came to blood pressure tracking, just 11% captured this data with a digital app or journal. However, this may reflect the higher-than-average of those diagnosed with elevated pressures, a senior population with a lower level of tech sophistication.

Lest all of this sound intimidating, there’s at least some good news here. Apparently, a full 86% of respondents said that they’d be willing to share data with their physician, a much larger share than those who would exchange data with a health plan (58%) or pharmacy (52%). In other words, they trust you, which is a big asset under these circumstances.

If you want to dive into digital health more deeply, here’s a few obvious places to start:

  • Link in-person and telemedicine visits: Rock Health found that a whopping 92% pf respondents who had an in-person visit first were satisfied with their video visit.
  • Be vigilant about data security: Almost 9 out of 10 consumers participating in the survey said that they would be willing to share data with you. Don’t lose that trust to a health data breach; it will be hard if not impossible to get it back.
  • Bring chronically-ill seniors on board: While this group may not be terribly inclined to digitize their healthcare, doing so can help you treat them more effectively, so you’ll probably want to make that point up front.

Like it or not, wearables, fitness bands, mobile health apps, and other digital health tools have arrived. It’s no longer a matter of if you take advantage of them, but when and how. Don’t be the last practice in your neighborhood that just doesn’t get it.

It’s Time To Work Together On Technology Research

Posted on September 12, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

Bloggers like myself see a lot of data on the uptake of emerging technologies. My biggest sources are market research firms, which typically provide the 10,000-foot view of the technology landscape and broad changes the new toys might work in the healthcare industry. I also get a chance to read some great academic research, primarily papers focused on niche issues within a subset of health IT.

I’m always curious to see which new technologies and applications are rising to the top, and I’m also intrigued by developments in emerging sub-disciplines such as blockchain for patient data security.

However, I’d argue that if we’re going to take the next hill, health IT players need to balance research on long-term adoption trends with a better understanding of how clinicians actually use new technologies. Currently, we veer between the micro and macro view without looking at trends in a practical manner.

Let’s consider the following information I gathered from a recent report from market research firm Reaction Data.   According to the report, which tabulated responses from a survey of about 100 healthcare leaders, five technologies seem to top the charts as being set to work changes in healthcare.

The list is topped by telemedicine, which was cited by 29% of respondents, followed by artificial intelligence (20%), interoperability (15%), data analytics (13%) and mobile data (11%).

While this data may be useful to leaders of large organizations in making mid- to long-range plans, it doesn’t offer a lot of direction as to how clinicians will actually use the stuff. This may not be a fatal flaw, as it is important to have some idea what trends are headed, but it doesn’t do much to help with tactical planning.

On the flip side, consider a paper recently published by a researcher with Google Brain, the AI team within Google. The paper, by Google software engineer Peter Lui, describes a scheme in which providers could use AI technology to speed their patient documentation process.

Lui’s paper describes how AI might predict what a clinician will say in patient notes by digging into the content of prior notes on that patient. This would allow it to help doctors compose current notes on the fly.  While Lui seems to have found a way to make this work in principle, it’s still not clear how effective his scheme would be if put into day-to-day use.

I’m well aware that figuring out how to solve a problem is the work of vendors more than researchers. I also know that vendors may not be suited to look at the big picture in the way of outside market researcher firms can, or to conduct the kind of small studies the fuel academic research.

However, I think we’re at a moment in health IT that demands high-level research collaboration between all of the stakeholders involved.  I truly hate the word “disruptive” by this point, but I wouldn’t know how else to describe options like blockchain or AI. It’s worth breaking down a bunch of silos to make all of these exciting new pieces fit together.

2019 CPT Codes To Cover Remote Monitoring And Digital Care Coordination

Posted on September 10, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

The American Medical Association has released CPT code set changes 2019, and among them are some new options specific to digital health practices.

While providing such codes is a no-brainer — and if anything, the AMA is late to the party – it’s still a bit of noteworthy news, as it could have an impact on the progress of digital care.  After all, the new codes to make it easier to capture the value of some activities providers may be self-funding at present. They can also help physicians track the amount of time they spend on remote monitoring and digital care coordination more easily.

The 2019 release includes 335 changes to the existing code set, such as new and revised codes for adaptive behavior analysis, skin biopsy and central nervous system assessments. The new release also includes five new digital care-related codes.

The 2019 code set includes three new remote patient monitoring codes meant to capture how clinicians connect with patients at home and gather data from care management and coordination, and two new “interprofessional” Internet consult codes for reporting on care coordination discussions between a consulting physician and the treating physician

It’s good to see the AMA follow up with this issue. To date, there have been few effective ways to capture the benefits of interactive care online or even via email exchanges between physician and patient.

As a result, providers have been trapped in a vicious circle in which virtual care doesn’t get documented adequately, payers don’t reimburse because they don’t have the data needed to evaluate its effectiveness and providers don’t keep offering such services because they don’t get paid for performing them.

With the emergence of just five new CPT codes, however, things could begin to change for the better. For example, if physicians are getting paid to consult digitally with their peers on patient care, that gives vendors incentives to support these activities with better technology. This, in turn, can produce better results. Now we’re talking about a virtuous circle instead.

Obviously, it will take a lot more codes to document virtual care processes adequately. The introduction of these five new codes represents a very tentative first step at best. Still, it’s good to see the AMA avoid the chicken-and egg-problem and simply begin to lay the tracks for better-documented digital care. We’ve got to start somewhere.

 

AI-Based Tech Could Speed Patient Documentation Process

Posted on August 27, 2018 I Written By

Anne Zieger is veteran healthcare consultant and analyst with 20 years of industry experience. Zieger formerly served as editor-in-chief of FierceHealthcare.com and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also contributed content to hundreds of healthcare and health IT organizations, including several Fortune 500 companies. Contact her at @ziegerhealth on Twitter or visit her site at Zieger Healthcare.

A researcher with a Google AI team, Google Brain, has published a paper describing how AI could help physicians complete patient documentation more quickly. The author, software engineer Peter Lui, contends that AI technology can speed up patient documentation considerably by predicting its content.

On my initial reading of the paper, it wasn’t clear to me what advantage this has over pre-filling templates or even allowing physicians to cut-and-paste text from previous patient encounters. Still, judge for yourself as I outline what author Liu has to say, and by all means, check out the write-up.

In its introduction, the paper notes that physicians spend a great deal of time and energy entering patient notes into EHRs, a process which is not only taxing but also demoralizing for many physicians. Choosing from just one of countless data points underscoring this conclusion, Liu cites a 2016 study noting that physicians spend almost 2 hours of administrative work for every hour of patient contact.

However, it might be possible to reduce the number of hours doctors spend on this dreary task. Google Brain has been working on technologies which can speed up the process of documentation, including a new medical language modeling approach. Liu and his colleagues are also looking at how to represent an EHR’s mix of structured and unstructured text data.

The net of all of this? Google Brain has been able to create a set of systems which, by drawing on previous patient records can predict most of the content a physician will use next time they see that patient.

The heart of this effort is the MIMIC-III dataset, which contains the de-identified electronic health records of 39,597 patients from the ICU of a large tertiary care hospital. The dataset includes patient demographic data, medications, lab results, and notes written by providers. The system includes AI capabilities which are “trained” to predict the text physicians will use in their latest patient note.

In addition to making predictions, the Google Brain AI seems to have been able to pick out some forms of errors in existing notes, including patient ages and drug names, as well as providing autocorrect options for corrupted words.

By way of caveats, the paper warns that the research used only data generated within 24 hours of the current note content. Liu points out that while this may be a wide enough range of information for ICU notes, as things happen fast there, it would be better to draw on data representing larger windows of time for non-ICU patients. In addition, Liu concedes that it won’t always be possible to predict the content of notes even if the system has absorbed all existing documentation.

However, none of these problems are insurmountable, and Liu understandably describes these results as “encouraging,” but that’s also a way of conceding that this is only an experimental conclusion. In other words, these predictive capabilities are not a done deal by any means. That being said, it seems likely that his approach could be valuable.

I am left with at least one question, though. If the Google Brain technology can predict physician notes with great fidelity, how does that differ than having the physician cut-and-paste previous notes on their own?  I may be missing something here, because I’m not a software engineer, but I’d still like to know how these predictions improve on existing workarounds.