.ppt

457 views

Published on

Published in: Health & Medicine
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
457
On SlideShare
0
From Embeds
0
Number of Embeds
3
Actions
Shares
0
Downloads
3
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • US spends $ 6-12bln per year to transcribe medical dictations into text
    Source: Medical Transcription Industry Association, MTIA
    195,000 death each year due to medical errors
    Unavailable patient information and illegible handwriting lead to diagnosing and ordering errors.
    Source: Health Grades Inc. "Patient Safety in American Hospitals," July 27, 2004
    Centralize IT administration to allow automatic updates, maintanance, monitoring
    Ban on handwritten documents
    Secure archiving
    Data acessability & encryption
    Reducing medical errors through better documentation.Edwards M, Moczygemba J.Health Information Management Program, Texas State University-San Marcos, USA.Preventable medical errors occur with alarming frequency in US hospitals. Questions to address include what is a medical error, what errors occur most often, and what solutions can health information technologies offer with better documentation. Preventable injuries caused by mismanagement of treatment happen in all areas of care. Some result from human fallibility and some from system failures. Most errors stem from a combination of the two. Examples of combination errors include wrong-site surgeries, scrambled laboratory results, medication mishaps, misidentification of patients, and equipment failures. Unavailable patient information and illegible handwriting lead to diagnosing and ordering errors. Recent technology offers viable solutions to many of these medical errors. Computer-based medical records, integration with the pharmacy, decision support software, Computerized Physician Order Entry Systems, and bar coding all offer ways to avoid tragic treatment outcomes. Persuading and training hospital staff to use the technology poses a problem, as does budgeting for the new equipment. However, the technology would prove its worth in time. The Institute of Medicine and coalition groups such as Leapfrog Group have recognized the problem that permeates the health care industry, manifests in many ways, and requires the many solutions that information technology offer.
  • Following are the complete results of Health Data Management's 2003 CIO Survey. The 25-question survey was sponsored by a grant from GE Medical Systems Information Technologies. Conducted in January 2003, the survey was mailed twice to 2,000 Health Data Management subscribers who are CIOs at hospitals or integrated delivery systems. It was e-mailed twice to 6,767 Health Data Management subscribers who are CIOs or directors/managers of data processing/I.T. at hospitals and integrated delivery systems. There were 804 responses, making this one of the largest surveys of its kind in the field
    http://www.healthdatamanagement.com/html/CIOSurvey.cfm
  • And Top business issues:
    http://www.himss.org/2005survey/healthcareCIO_final04.asp
  • Impact of Poor Documentation
    Less reimbursement means less revenue
    Less capital allocation for new equipment
  • Here we can see the history of automated document creation
    We started with discrete speech recognition.
    In fact Philips never participated there because we knew that [discrete] [speech] [recognition] [would] [never] [be] [used] [by] [professional] [users]
    Then we introduced the world’s first continuous speech recognition product which was called SP6000 at these days.
    It even had its own hardware!
    However continuous speech recognition means still that you have to dictate word by word what was said. Otherwise you lose time and money for correction.
    In 2001 Philips started it’s first initiative of ISI (Intelligent Speech Interpretation)
    We introduced auto-punctuations, auto-formattings and headers to make it more natural.
    And we are on the way to further improve ISI until it will be completely free form input with structured output
    People will be able to simply dump their brain.
    No matter about the wording.
    No matter about the formattings.
    Just the information counts.
    We are not their yet.
    Authors still need to follow some rules to get the best out of the tool.
    But we expect to be so far in 5-10 years that we can confidently say “People can express themselves freely”
    You might say: This is a nice picture but where are we today?
  • <number>
  • <number>
  • Prediction:
    Paper, manila notes, and folders will become memorabilia of the past and relegated to museums
    Payoff includes decreasing receivables, improving documentation, reducing documentation time and getting data
  • .ppt

    1. 1. Speech Recognition Aids Faster Access to Clinical Information Nick van Terheyden, MD – Chief Medical Officer, Philips Speech Recognition Systems August 7, 2006
    2. 2. 2 Physician’s Nightmare • Mountains of paperwork • Patients in need of attention • Mountains of paperwork • Clinical staff in need of answers • Mountains of paperwork • Personal need for rest, family time and reflection to be a better practitioner
    3. 3. 3 Complex process of caring for patients • Clinical documentation comprises the essential building blocks for all healthcare processes • Obtaining a history • Performing a physical examination • Ordering tests and evaluating their results • Establishing diagnoses • Prescribing therapies and monitoring the course of treatment. • Each step must be accurately and thoroughly documented
    4. 4. 4 My Dream One step process to document patient care accurately • Ability to focus on patient care • Allow me to discuss, respond and instruct the clinical team appropriately • Facilitate personal needs for R&R
    5. 5. 5 Health Data Management 2003 CIO Survey When doctors and nurses at your organization express dissatisfaction with internal services, which ONE of the following is their most frequently expressed concern? 59% 0 0.1 0.2 0.3 0.4 0.5 0.6 Long working hours Dissatisfaction with IT structure Too much paperwork Difficulty maintaining patient satisfaction levels Lost income due to claims
    6. 6. 6 HIMSS 2005 Survey Technology Adoption in the next two Years
    7. 7. 7 Volume and Complexity Increasing • Tremendous increase in volume and complexity of studies in the US during the past several years – “2005-2010 workload to increase 50% for Radiologists” James Thrall, Mass. General Hospital, Boston – Rick Marin, Mayo Clinic Jacksonville, FL – 2002 16,000 CT images/day (2 sec/image) – 2006 80,000 CT images/day (0.45 sec/image) • In the Next 10 years we will learn/develop more medical information than we have captured in the whole of medical history No of Radiology Studies 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 2002 2003 2004 2005 2006 2007 2008 2009 2010
    8. 8. 8 Documentation Needs to Support the Right Clinical Decisions• Clinical medicine is inherently challenging – Complexity of the human body – How it responds to disease • Our understanding of illnesses and optimal treatment is continually changing • AND doing so faster than ever • Clinical decisions are fundamentally considered judgments
    9. 9. 9 Physicians Are the Drivers • Only physicians can document diagnosis and procedures • Possible to document the clinical course, but – Fail to provide accurate, specific documentation – Complexity, co-morbidity, complications • Diagnosis and Procedures are required for reimbursement of care delivered
    10. 10. 10 © Gartner Hypecycle for Healthcare Provider Technologies 2006
    11. 11. 11 Speech Recognition: Where are We and Where are we Going The evolution of automated document creation Time Ease of Use Free Form Dictation (Discrete SR) Continuous SR Intelligent Speech Interpretation 1990 1994 2001 2010 - 20152005 Interactive
    12. 12. 12 Clatterbridge Center for Oncology (CCO) • One of the largest cancer treatment facilities in the United Kingdom • Diagnostic Imaging – 19,000 exams per year – X-Ray – Ultrasound – CT – Nuclear Medicine – MRI • Target of Report Available within 24 hours
    13. 13. 13 Report Card For CCO • 1 in 4 MRI’s met deadline – available within 24 hours • Total of 11% of imaging reports available within 24 hours • Frustrated Users • Frustrated Referring Physicians • Frustrated patients
    14. 14. 14 Availability of medical reports within 24hrs Report type 2003 CT 41% MRI 25% Nuclear Medicine 43% X-Ray 11%
    15. 15. 15 Report type 2003 2006 CT 41% 86% MRI 25% 76% Nuclear Medicine 43% 72% X-Ray 11% 80% Availability of medical reports within 24hrs
    16. 16. 16 “Speech recognition didn’t change my working procedure. What it did change though is report turnaround.” Dr. Keith Grant Clinical Director of Diagnostic Imaging Other Benefits Improved Data security • No stray tapes to loose • Imaging reports recorded directly into patient record No misfiling • Report demographics data tied directly at the time of report dictation • Instant report retrieval
    17. 17. 17 Resistance • Secretarial Support – potential job cuts • Physician resistance to change – Providing a practice environment • Technology mistrust – there was something secure in being able to hold onto a physical tape of a report • Rotating Staff – enrollment and training time • Variation in Phrases • Redundant Re-Working
    18. 18. 18 Report Standardization Dictated Words Heading per departmental guidelines To Conclude Conclusion CT Scan of the abdomen prone IV CT Abdomen (prone + IV) Dynamic CT Scan CT Dynamic CT Scan of the neck, chest and liver CT Neck, Chest, Liver CT scan of the chest with intervenous contrast CT Chest (with IV contrast) “Our dedication paid off quickly, as the system now not only recognizes words but interprets them according to our clinic’s guidelines.“ Anne-Marie Aspinal Clinical EPR Trainer
    19. 19. 19 Dictated Text Recognized Text Standard phrases End of dictation. Thank you. Does not appear in final document. Redundant phrases Send copy of report to Does not appear in final document. Section headings Condition on/at/upon/of/ discharge Next is condition... Next section is condition.. Condition on Discharge Dates Fifth of May five two thousand two May fifth two thousand six... 5 May 2006 Automatic punctuation No chills fevers night sweats weight loss... No chills, fevers, night sweats, weight loss... Silences/pauses There has not been (------pause----) much change... There has not been much change. Non-speech dictation There has not been (paper rustling) much change... There has not been much change... Hesitations There has not been (AAHHMMMMM) much... There has not been much... Contraction There hasn‘t been There has not been Orthographic variants She is here today because she has a letter from her (GP) resident from the university... She is here today because she has a letter from her GP or General Practitioner or Family Practitioner (customizable) … Understanding more... Intelligent Speech Interpretation
    20. 20. 20 Critical Success Factors • Accept resistance to change and work with clinicians to overcome using multiple approaches – Physician champions – Test environment – Capability to move from front end to back end seamlessly • Start with the low hanging fruit – Implement back end which requires minimal change in behavior – Leave option available to move to front end when the time is right • Integrate the technology into the Picture Archiving Communication System (PACS) or Radiology Information System (RIS) – Islands of technology create support nightmares
    21. 21. 21 Speech Recognitions Impact to Medical Documentation • Speech recognition can reduce costs by 30-40% • Early users will have high competitive advantage • Question is not if, but when you should use Speech Recognition
    22. 22. 22 So What Does the Future Hold •Medical Technology is fragmented and increasingly complex •There are many developments on the Horizon •Many inputs to the medical record •Massive Legacy investments that institutions and individuals are either unwilling or unable to migrate •It is not going to get any Easier!
    23. 23. 23 Conclusion• To achieve our goals – Capture the data at source to feed the EMR – Support clinical decision making with complete codifiable data – Error reporting needs to become error catching • Speech is the most natural form of communication • Speech Recognition bridges the gap between clinicians and technology • SpeechMagic no longer “nice to have” this is a “must have technology”
    24. 24. 24 http://www.philips.com/speechrecognition For further information please contact: Nick van Terheyden, MD Chief Medical Officer 888-speak-50

    ×