Speech Recognition

In a freshly inked article from Health Imaging & IT, healthcare IT executives and vendors unveil their respective visions and roadmaps for speech recognition. Integration clearly is the main course on the menu. While this is too bad for those vendors who think HL7 is the name of a 1990’s boy band, this article further confirms hospitals’ appetite for structured documentation with a direct impact on patient care.

• Dr. Stephen Rosenthal, M.D. from the Jewish General Hospital in Montreal, QC, insists on the importance of interfaces between speech recognition and third-party systems in order to deliver the foundation for evidence-based medicine through searchable, standardized clinical data: “if speech systems aren’t relatively uniform, people will find systems on their own and use them. Then you have a hodgepodge of systems that don’t talk to each other and standardization is lost. We are much better investing in something uniform.”
• According to Terence Matalon, MD, from the Albert Einstein Medical Center in Philadelphia, PA, speech recognition & PACS integration is a must have for Radiologists since it eliminates the need to re-enter patient information or “have two applications open to attain the same goal”. Matalon even pushes the point further by making integration expertise a competitive differentiator between vendors: “there are dozens of products that can reliably show you the current exam, prior exam and reports. The differentiating factor is how well they integrate with third parties and how well they reduce amount of work involved in interpreting reports and generating reports.”
• On the vendor side, Klaus Stanglmayr from Philips Speech Recognition Systems explains how “interoperability and the ability to exchange data between systems and countries is becoming more and more critical in Europe. Standardized terminology would prevent the need to have data translated from one language to another.”
• Finally, Chris Spring from MedQuist insists on vendors’ primary mission to “make it easier for the physician to accept the technology.”

Now what’s your vision? Share it on this blog!

In a recent article from Health Imaging & IT, the respective Chief Medical Officers of Philips Speech Recognition Systems and Health Language shed light on the future of speech recognition technology for healthcare. Beyong speech-to-text conversion, SR is clearly turning into a full enterprise solution with a direct impact on both clinical decision-making and the actual cost of healthcare.

Nick van Terheyden, MD, CMO, Philips Speech Recognition Systems:

Speech recognition can reduce costs by 30 to 40 percent, and early users will have a very high competitive advantage.

Providing clinically actionable data is the key to solving fundamental challenges with EMRs such as: capturing data at the source to input into the EMR, supporting clinical decision-making with clinically actionable data, and providing tools that enable the capability to catch errors before they’re committed.

Speech recognition and natural language understanding bridges the gap between clinicians and technology

Brian Levy, MD, CMO of Health Language:

The language engine works simultaneously with the speech engine allowing for real-time conversion of text to standards. In addition, an EMR may be set up to allow smaller sections of dictations for various components of the record such as history, problem lists, medications, and so on. The EMR or other application may then permit the clinician to select more specific or additional codes to further standardize the information.

The current interaction with the patient usually occurs separately from the ability to look up referential material pertinent to the patient. But the introduction of speech and conversion to standards within an electronic medical record enables real-time information.

> Read full article

In a recent article published by US magazine For The Record, speech recognition is listed as one of the hottest issues for 2008 by healthcare industry leaders themselves. Here’s a sneak peek:

Brenda J. Hurley, CMT, FAAMT, director of industry relations and compliance at Medware

Institutions and medical transcription services will continue looking at speech recognition (SR) technology in 2008. In the transcription world, that’s OK. We look to use technology for better efficiency. But don’t put transcriptionists in a world where we’re just correcting the same errors every day and then expect increasing productivity.

For the new year, Hurley is hoping to see SR implementation that incorporates two main cost-effective strategies: standardization of healthcare documents and following dictation best practices format. By applying these two factors, she believes greater efficiency will be achieved with traditional transcription, as well as the technological investment. “Why wait? Prepare now for technology to move ahead. Tackle the basics, and meanwhile, you can reap the benefits.”

Amber Doster, vice president of marketing at HealthPort:

Technology will be a valuable tool in the hands of healthcare organizations. More and more, we are seeing that providers want access to an increasing detail of information. In 2007, we experienced a spike in requests for product functionality and services that centered around pulling information from areas that have never before been accessed. Paper records don’t just need to be in the provider’s hands, they need the ability to quickly search through many types of documents for certain terms or procedures for patient care. In 2008, this need will only grow, and it will help drive solution development from a vendor perspective.

> Read full article

Who would have thought so just 3 or 4 years ago? According to a scientific presentation at RSNA 2007, the annual meeting of the Radiological Society of North America, transcribed reports show higher error rates than the ones returned from speech recognition applications. The study, conducted at the Radiology Consultants of Iowa, a non-academic radiology group in Cedar Rapids, Iowa, compiled error rates for 498 reports created with a system powered by SpeechMagic and compared those with error rates for the same reports transcribed in a traditional manner.

The traditionally transcribed reports included at least one error in 13 percent of the total, while the speech recognition reports demonstrated one error in only 9 percent of the total studies.

The rate for significant errors, requiring the preparation of an addendum, was 0.6 percent for speech recognition and 2 percent for traditional transcription,” reports Floyd, RCI partner.

He shared that the accuracy rate for speech recognition reported by RCI was confirmed by an independent analysis conducted at one of the two acute care hospitals that the group services. The facility’s evaluation of 514 reports conducted in September produced an overall transcription error rate of 9.7 percent for both automated and traditional report generation, and that an addendum was required for 0.6 percent of the reports.

The report cohort consisted of 20 to 25 studies involving CR, MR, and general radiographic procedures from each of the 24 radiologists in the practice.

> Read full article

Here is a Time Magazine article from earlier this year where author Jeremy Caplan digs further into the preventable medication mistakes statistics. Caution, coarse figures ahead…

Doctors’ sloppy handwriting kills more than 7,000 people annually. It’s a shocking statistic, and, according to a July 2006 report from the National Academies of Science’s Institute of Medicine (IOM), preventable medication mistakes also injure more than 1.5 million Americans annually. Many such errors result from unclear abbreviations and dosage indications and illegible writing on some of the 3.2 billion prescriptions written in the U.S. every year.

“Thousands of people are dying, and we’ve been talking about this problem for ages,” says Glen Tullman, CEO of Allscripts, a Chicago-based health care technology company, that initiated the project. “This is crazy. We have the technology today to prevent these errors, so why aren’t we doing it?”

Although some doctors have been prescribing electronically for years, many still use pen and paper. This is the first national effort to make a Web-based tool free for all doctors. Tullman says that even though 90% of the country’s approximately 550,000 doctors have access to the Internet, fewer than 10% of them have invested the time and money required to begin using electronic medical records or e-prescriptions.

SureScripts CEO Kevin Hutchinson says one key to reducing medication errors is to get the most prolific prescribers to transition to electronic processing. “Not a lot of people understand that 15% of physicians in the U.S. write 50% of the prescription volume,” Hutchinson says. “And 30% of them write 80%. So it’s not about getting 100% of physicians to e-prescribe. It’s about getting those key 30% who prescribe the most. Then you’ve automated the process.”

> Read full article

In this month’s issue on Health Imaging & IT, Nick van Terheyden, Chief Medical Officer for Philips Speech Recognition Systems, further comments on the boom of medical knowledge and the market pressures leading healthcare facilities to seriously consider the speech recognition option:

Speech Recognition: a competitive advantage

The addition of speech recognition technology (SR) to (RIS and PACS) systems can further reduce practice overhead, while providing a competitive advantage to the groups that implement them. Speech recognition can reduce costs by 30 to 40 percent, and early users will have a very high competitive advantage.

Transcription: major costs for hospitals

As exam volume continues to grow, transcription costs will match pace; although reimbursements for the performed procedures may not. According to van Terheyden, approximately $12 billion is spent annually in the U.S. on medical transcription, making it one of the top five line-item costs of hospitals and healthcare systems.

“I think the question for everybody is not if, but when you should use speech recognition,” van Terheyden said.

5 goals for automating clinical documentation

van Terheyden outlines five goals for automating clinical documentation:
1. Reduce physician time documenting
2. Improve availability of documentation
3. Re-use standard phraseology
4. Improve quality and consistency of documentation
5. Decrease the cost of documentation

The US: largest speech recognition user in the world

There are approximately 40,000 active physician users generating about 18 million lines per month with SR technology. As the need to reduce costs in healthcare delivery accelerates, this user base will increase.

More medical information in the next 10 years than in the whole medical history…

The next 10 years will see more medical information learned and developed than has been captured in the whole of medical history. Radiology, in particular, is bearing the lion’s share of this information glut. James Thrall, MD, of Massachusetts General Hospital in Boston projects that the radiology workload will increase approximately 50 percent by 2010.

> Read full article

In a recent interview with a UK healthcare IT publication, Philips Speech Recognition Executives comment on a couple of mind-blowing figures.

Dr. Nick van Terheyden, Chief Medical Officer:

Clinical knowledge is estimated to double every 18 months. Given the massive volume of medical research, it is impossible for any physician to keep up with all new developments. A new study on a specific condition or its treatment might have been published just the day before the physician sees a patient.

Physicians are trained to identify conditions based on the occurrence of signs and symptoms. But as medicine expands and the number of diagnoses increases it becomes harder to use this information to identify specific conditions and their cause. As medicine is undergoing an explosion of information, technology must be utilized to assist the clinical team in prioritizing information, highlighting key data and guiding the care process.

…If the EHR is able to provide real-time access to clinical data the physician becomes much more of a pilot, navigating and taking decisions as critical and relevant information is presented to him in a clear and effective way.

Marcel Wassink, CEO:

A survey by the European Commission found that almost 4 in 5 EU citizens classified medical errors as an important problem in their country. Other studies suggest that:

• In Italy, up to 90 people die per day because of an error
• In Germany this number is estimated at 38,000 per year.
• In the UK, 850,000 medical errors are reported each year.
• In the US, it is estimated that 100,000 people die from medical errors each year.

First measurements show that a hospital’s overall productivity can be increased by 5-7% just by increasing reporting efficiency. Using speech recognition to improve reporting accuracy is next, along with ensuring faster availability and accessibility to information. I see this as the basis for reducing the alarming error rates in healthcare.

The feedback we receive from doctors says that the length of their reports has increased, since they started using speech recognition – because it makes reporting much easier. Today, doctors are often advised to keep their dictations short, to ease the workload on the transcriptionist. This clearly can’t be the solution – not in a time where we are moving towards evidence-based medicine. More information is what doctors need and that’s why speech recognition must be offered to doctors as a standard option for capturing information electronically, the same as the keyboard or the mouse.

The above figures are the reality check that motivated the development of Philip’s Interop, a concept demonstrator designed to advance system interoperability in healthcare with the help of speech recognition, Map of Medicine (Medic to Medic), Health Language and Elsevier technologies.

In the latest issue of For the Record Magazine, Robbi Hess discusses the pros and cons of the Once-and-Done transcription model (the other name for front-end speech recognition, whereby physicians see the results of their dictation on screen and make their own corrections). First, the article highlights findings from a Gartner Report by Barry Hieb, MD, healthcare research, titled “The Evolving Model of Clinical Dictation and Transcription:”

The lack of efficiency and the money that leaks through transcription cracks have always been issues in the healthcare industry. The role of dictation and transcription in clinical documentation is evolving in response to new technologies and new functional requirements…

Traditional dictation and transcription are giving way to ‘editor-based’ approaches and that once-and-done dictation will eventually be adopted in the majority of situations.” [...] Because speech recognition makes increased productivity and associated cost savings possible, it is now an integral part of most new dictation and transcription contracts.

The OAD dictation model will take longer to unfold but will be driven by the need to provide value back to physicians at the time they are dictating reports.

The importance of the physician’s sign off
The Author then quotes Peter Preziosi, PhD, CAE, CEO of the Medical Transcription Industry Alliance and executive director of the Association for Healthcare Documentation Integrity (AHDI):

The reality today is that you have physicians that don’t even sign off on charts even though they are legally liable for the content. My concern is that when we look at building a national healthcare information infrastructure, it will be that much more critical to ensure the accuracy and completeness of information. The onus of chart accuracy must be on the clinician, not the MT, making the need for clinician sign-off imperative.

Back-end vs Front-end Speech Recogniton
Hieb says using speech recognition technology cuts significant costs from the entire transcription cycle:

The editors can crank out 50% to 100% more copy a day and, as a result, the hospital gets charged less money. But the downside is there is still a two-day turnaround time, and the hospital is still paying for both transcription and dictation costs.

The advantage to the editor mode of transcription is that the doctor is not being asked to change the way he or she operates, and the report is turned around faster. Presumably, the editor is happier, is doing more work, and is being more productive. And from a documentation standpoint, you are telling the physician that the report will be back more quickly, but you still have to look at it, revise it, and sign off on it.

The OAD model, carries with it a good news/bad news scenario: The hospital can save money and enhance performance, but the doctor has to change his or her dictation routine.

Doctors are hesitant now because we will be telling them, ‘You will be dictating at a computer, but you can see what you are dictating’. Hieb acknowledges that some doctors are poor dictators, but with OAD, they can receive direct feedback and make edits as they go, while the patient information is fresh in their minds.

What the future holds…
Hieb believes that one of OAD’s benefits is that when the clinician dictates, edits, and signs off on the record, it’s ready to go into the electronic chart:

The turnaround time has dropped from four days to two days [with back-end speech recognition] to two minutes, and now any doctor can see that report as soon as the physician signs off on it. That turnaround time brings nothing but benefits and better care to the patients.

Although a hospital would have to invest in the software and hardware technologies necessary to implement an OAD system, Hieb says those expenses would pay off in the long run. “There will be set-up and maintenance fees, but they will be nowhere near the costs of the money spent on dictation and transcription.” He agrees, though, that getting doctors to change their behavior will be the largest hurdle to overcome.

With OAD, there is no subsequent time added to that chart, little hassle, and minimal risk of error. The single best defense against malpractice is good documentation, and with once-and-done, you have given the doctor the chance to do the reports and be done with it.

How physicians are accepting speech recognition technology…
Hieb goes on to comment:

Doctors are surprised to see that they can save time and money. If they spend a little bit of time up front ‘training’ the system, they are reaping dividends in time and money saved.”[...] OAD is being accepted more readily in private practices. That is where the technology is really making inroads because the doctors see they can save time and money, and if they have an electronic copy of the record, the staff isn’t busy chasing down records. In fact, they may be able to reduce the amount of staff they are paying. The OAD knows to file the record in Susie Smith’s chart, and the general trend in medicine is toward more clinical automation.

OAD can also be effective in the emergency department (ED), where time is of the essence. In the ED, the real benefit is getting that data out there and into the chart instantly. The more quickly and effectively the information is captured in the chart, the more quickly the physicians have access to that data.

Conclusion

For Hieb, the reason to embrace OAD is because the goal of the healthcare system is to help sick people get well and healthy people stay healthy. “We are entering an age when information is a critical component of achieving these two goals, and once and done is a better, more efficient way to capture that information,” he says.

Preziosi says the concept is not feasible in today’s marketplace, even with the enabling technologies that will be seen in the future. “Given the cost restraints, the persistent labor shortages, and the increased demands on the healthcare system, I don’t see OAD as being realistic,” he explains. “I think the clinical documentation sector needs to listen to the concerns of the consumers of our services and adapt our service offerings to meet their ever-evolving demands.”

> Read full article

Not much to comment on when figures speak for themselves. According to the latest Gartner report, healthcare IT spending is still on the rise with healthcare facilities’ budgets for business processing and IT services to show a 23% growth by 2011. Between 2006 and 2007 alone, facility contracts for information systems have increased by 13%:

The healthcare provider industry (worldwide) will spend $77.3 billion in 2007 on IT (hardware, software, IT services, internal services and telecommunications). The compound annual growth rate (CAGR) from 2006 through 2011 will be 5.2%. Key growth areas in 2007 will be software and services.

Shifting money figures usually reflect a more profound, behavioural change. And here we go:

In 2005, 8% percent of healthcare facilities were considered early adopters of health IT, while last year, 33% of facilities were considered early adopters, which can be attributed to changes in how executives perceive the need for information technology, according to John Lovelock, analyst and director of healthcare research at Gartner.

Fall 2007 is definitely harvest season for market intelligence in our sector of interest: healthcare IT. After the MRI’s and HealthImaging Surveys, now comes the HIMSS 18th Annual Leadership Survey results. The survey “reports the opinions of information technology (IT) executives from healthcare provider organizations across the U.S. regarding the use of IT in their organizations. The study was designed to collect information about IT priorities, technology adoption, application usage and other crucial factors in the use of IT to enhance healthcare.” Here are a few interesting findings…

IT priorities

Implementing technology to reduce medical errors and to promote patient safety continues to be a top priority (35%), both now and for the future. This is being driven by a focus on quality of care and patient satisfaction, which were identified most frequently as the healthcare business drivers having the most impact in the next two years.” Workflow redesign still scores 30%.

Technology adoption

Speech recognition is on the list of technologies that survey respondents intend to implement in the next two years, although Intranet and high-speech networks have become the number one priorities:

Security technologies

Electronic signature appears as one of the top technologies to answer respondents’ primary concern regarding the security of data at their organization:

IT budgets

Nearly threequarters of respondents reported that their IT operating budget will increase in the next year. Many respondents attributed this to an overall growth in the number of systems and technologies.

Vendor satisfaction

In general, respondents were satisfied with the overall IT products/services they receive from suppliers, application vendors and consulting firms—60 percent of respondents indicated that they were satisfied.

>> Read full Survey from the HIMSS web site