What is Biomedical and Health Informatics?

Transcription

1 What is Biomedical and Health Informatics? William Hersh, MD Professor and Chair Department of Medical Informatics & Clinical Epidemiology Oregon Health & Science University Portland, OR, USA Web: Blog: informaticsprofessor.blogspot.com 1

2 Overview Overview of discipline and profession Definitions of important terms Medical (or clinical) informatics Person-specific applications Knowledge-based applications Bioinformatics Education and training 2

3 A field with supporters in high places To improve the quality of our health care while lowering its cost, we will make the immediate investments necessary to ensure that within five years, all of America s medical records are computerized It just won t save billions of dollars and thousands of jobs it will save lives by reducing the deadly but preventable medical errors that pervade our health care system. - January, 2009 Health Information Technology for Economic and Clinical Health (HITECH) Act of American Recovery and Reinvestment Act (ARRA) of 2009 invested up to $30 billion in health information technology 3

4 Opportunities are not limited to healthcare Modern biomedical scientists use computers and robots to separate molecules in solution, read genetic information, reveal the three-dimensional shapes of natural molecules like proteins, and take pictures of the brain in action. All of these techniques generate large amounts of data, and biology is changing fast into a science of information management. There is no way to manage these data by hand. What researchers need are computer programs and other tools to evaluate, combine, and visualize these data. - Vision for integration into healthcare: Stead,

5 Other areas of opportunity Public health protecting the public and promoting health e.g., disease surveillance, such as for H1N1 or bioterrorism Consumer health enabling better management of health Personal health records for engaged patients and consumers Imaging use of images and their analysis for biomedical research and clinical care And more 5

6 It is one of ten ahead of the curve careers Careers that are relatively new, already viable, and promise further growth (Nemko, 2007) Biomedical and Among Should have leading said: centers: University Oregon Health of Oregon & Medical Science University Center 6

7 What is biomedical and health informatics? I get asked this so often that I keep a Web site I have also written about it Overview of medical informatics (Hersh, 2002) But there are barriers (Hersh, 2004) Characterization of and changes in the profession (Hersh, 2006) Many career opportunities as well (Hersh, 2008) Definitions of field (Hersh, 2009) Workforce needs (Hersh, 2010) 7

8 Let us start by defining informatics Field concerned with how people use information, usually aided by technology, to improve aspects of the world More about information than technology, but technology is essential The science of health service design must be a science of sociotechnical systems, and today that science is called informatics. (Coiera, 2007) (SUNY Buffalo) 8

9 It has a fundamental theorem (Friedman, 2009) Goal of informatics is: and not: 9

10 My current preferred terminology (Hersh, 2009) Biomedical and health informatics (BMHI) is the field concerned with the optimal use of information, often aided by technology, to improve individual health, healthcare, public health, and biomedical research Differs from information technology (IT) in that Is strongly rooted in domains (e.g., health) IT is one (of many) tools employed Practitioners of BMHI are usually called informaticians (sometimes informaticists) Disagreements over terminology in both noun and adjectives preceding it Has an adjective problem 10

11 Categories of BMHI (Hersh, 2009) Imaging Informatics {Clinical field} Informatics Research Informatics Consumer Health Informatics Bioinformatics (cellular and molecular) Medical (Clinical) Informatics (person) Public Health Informatics (population) Legal Informatics Biomedical and Health Informatics Chemoinformatics Informatics = People + Information + Technology 11

12 Categories of BMHI (cont.) Bioinformatics application of informatics in cellular and molecular biology, often with focus on genomics Medical/clinical informatics application of informatics to individuals Informatics applied in a more specific healthcare domain is {X} informatics, e.g., nursing, dental, pathology, primary care, etc. Public health informatics application of informatics in public health Over-arching aspects of BMHI Imaging informatics focus on images in all categories Research informatics focus on research in all categories 12

13 Informatics now viewed as a core competency for health professionals According to Institute of Medicine report, the modern health professional must have competency in informatics as part of larger goal to provide patient-centered care (Greiner, 2003) Informatics competency is not just computer literacy! The Google generation (aka, digital natives ) does not necessarily have good information skills (CIBER, 2008) 13

14 Some historical perspective on informatics (Collen, 1994) Origin of term from Russia in late 1960s Achieved widespread use in France (informatique) and later rest of Europe in 1960s to denote computing issues related to information use Medical informatics first used in 1974 European perspective documented by Hasman (1996) and Moehr(2004) At present, most significant use is in biomedical arena, but it is used by other domains, such as law, chemistry, social sciences, etc. 14

15 How is informatics distinguished from related terms? Information technology (IT) computer and related technology Health information technology (HIT) healthrelated application of IT Computer science (CS) is academic discipline that underlies IT (and other technologies) Management information systems (MIS) is another field underlying IT (usually in business schools) 15

16 Other related terms Health information management (HIM) discipline historically focused on management of (paper) medical records (changing in current environment), with three main levels of practice Registered Health Information Administrator (RHIA) highest level, baccalaureate degree Registered Health Information Technologist (RHIT) associate degree Certified Coding Specialist (CCS) usually less then associate degree 16

17 Other related terms Information and communications technology (ICT) same as IT with added emphasis on telecommunications ehealth use of ICT for health Telemedicine provision of healthcare when participants separated by time and/or distance Sometimes applied in specific clinical specialties, e.g., teleradiology, teledermatology Telehealth pursuit of health when separated by time and/or distance 17

18 Other related terms Evidence-based medicine (EBM) the application of the best scientific evidence in the medical decisionmaking process Evidence-based practice (EBP) the application of EBM in medical practice Comparative effectiveness research (CER) research that compares one or more diagnostic or treatment options to evaluate effectiveness, safety, or outcomes Information retrieval the field devoted to searching (mostly text, mostly knowledge-based information) 18

19 Other terms central to medical/clinical informatics Electronic health record (EHR) patient s health record in digital form Has mostly supplanted the term electronic medical record (EMR) Personal health record (PHR) personally controlled health record, which may or may not be tethered to a healthcare organization s EHR Health information exchange (HIE) exchange of health information across traditional business and other boundaries Organization managing HIE used to be called a Regional Health Information Organization (RHIO) HITECH Act introduced a new type of organization Regional Extension Center (REC) 19

20 Other terms related to the EHR Clinical decision support (CDS) alerts, reminders, rules designed to improve clinician decision-making Computerized physician/provider order entry (CPOE) with or without CDS Interoperability and standards Privacy, confidentiality, and security 20

21 Research informatics Clinical research informatics (CRI) is area of informatics applied to clinical research (Embi, 2009) Increasing recognition that research findings must translate into clinical care more quickly and efficiently, leading to NIH investment in clinical and translational research through the Clinical & Translational Science Award (CTSA) program (Zerhouni, 2007) Difference between IT and informatics very evident in this domain (Bernstam, 2009) Translational bioinformatics (Sarkar, 2011) increased understanding of human genome may lead to more personalized medicine tailored from one s genome (Hamburg, 2010) 21

22 Major applications in medical/clinical informatics Based on two core types of information, with different uses and applications Person-specific information is generated in the care of patients Applications: electronic health records, telemedicine, etc. Knowledge-based information is the scientific literature of biomedicine and health Applications: information retrieval systems, evidencebased medicine 22

23 Person-specific informatics applications Electronic health record (EHR) EHR definitions and components Secondary use (re-use) of clinical data HIE Overcoming barriers and spurring adoption Telemedicine Provision of care over time or distance via communications technologies 23

24 Motivations why do we need more IT in healthcare? Quality Safety Cost Inaccessible information 24

25 Healthcare quality There are many studies to choose from McGlynn, 2003 Sample of nearly 7,000 adults in 12 US metro areas assessed for 30 conditions On average, only 54.9% of care was consistent with known quality NCQA, 2009 annual report on quality shows gaps to get all health plans to 90 th percentile of current quality 49, ,300 avoidable deaths $12 billion in avoidable medical costs Quality of care for patients with chronic disease no better and in many ways worse in US than for other developed countries (Schoen, 2009) 25

26 Safety and medical errors The IOM Errors report: As many as 98,000 Americans die each year due to medical errors, mostly medication errors (Kohn, 2000) Some have argued that the numbers are too high or too low, but none argue with the concept A decade later, problem persists (Van Den Bos, 2011) Lost in the discussion: Most errors are the result of faulty systems; the solution is not in making people smarter or punishing them, but building better systems to identify and prevent errors (Berwick, 2003) Medicine used to be simple, ineffective, and relatively safe. Now it is complex, effective, and potentially dangerous. (Chantler, 1999) 26

27 Cost Healthcare costs continue to rise and outpace inflation (Wilson, 2011) US spends more per capita on healthcare but gets less in terms of products (OECD, 2006; Angrisano, 2007) and outcomes (Banks, 2006; Lasser, 2006) 27

28 Inaccessible information Primary care physicians reported information missing in 13.6% of clinical visits (Smith, 2005) In 52% of instances, information was available but outside system In 44% of instances, lack of information could adversely effect patients As many as 20% of all tests and 1 in 7 hospital admissions may be result of inadequate access to information (David Brailer, unpublished data) HITECH investment provides opportunity to achieve the learning health system (Friedman, 2010) 28

29 EHR definitions key capabilities (IOM, 2003) Health information and data Result management Order management Decision support Electronic communication and connectivity Patient support Administrative processes Reporting and population health management 29

30 EHR data flow Additional financial and administrative data Departmental system Data Warehouse Departmental system Data Repository Departmental system Repository is usual place for EHR data to reside and be accessed Regional and national systems 30

31 Benefits and challenges of the EHR Benefits Improved physician, nursing, and other documentation and care Personal health records Clinical decision support Quality assessment Public health Clinical research Health information exchange Challenges Data quality Data usability Standards and interoperability Privacy, confidentiality, and security Understanding clinical narrative text Implementation 31

32 Clinical decision support (CDS) CDS uses EHR data to provide context-specific advice, alerts, and reminders, such as Assisting with choices in diagnosis and therapy Detecting problematic situations, such as medication errors or drug-drug interactions Types of CDS Information display showing information in context of situation Reminder systems reminding clinicians to perform actions Alerts alerting to critical clinical situations Though growing concern over alert fatigue Clinical practice guidelines guiding treatment to provide normalized care based on best evidence 32

33 EHRs allow and align secondary use (or re-use ) of clinical data Additional uses of EHR data include (Safran, 2007) Personal health records (PHRs) Clinical and translational research generating hypotheses and facilitating research Health information exchange (HIE) Public health surveillance for emerging threats Healthcare quality measurement and improvement One important tool for re-use of clinical data is natural language processing(nlp), which has been challenging but is seeing growing successes (Stanfill, 2010; Nadkarni, 2011; Chapman, 2011) 33

34 Example of an EHR Using the Veterans Health Information Systems and Technology Architecture (VistA) Why VistA? A state-of-the-art EHR that has transformed healthcare in the Veteran s Health Administration (VHA) (Brown, 2003; Greenfield, 2004) Not that pretty, but has all of the modern features of the EHR, e.g., clinical decision support (CDS), computerized provider order entry (CPOE), etc. Distributed under open-source model, unlike most other vendors who do not even allow screen shots to be shown outside their customers institutions 34

35 Some details about VistA Is available as a demo Demo version has following screens but not data Application has two components Server written in M (formerly called MUMPS), accessed via command-line interface Runs in commercial Intersystems Cache (on many platforms) or open-source GT.M (Linux only) Client written in Delphi and providers graphical user interface Only runs on Windows (just about all versions) 35

36 Cover sheet after patient selected

37 Drilling down to details of a problem

38 Details of an allergy 38

39 Viewing vital signs over time

40 More details on problems

41 Viewing the patient s notes

42 Adding a new note

43 Viewing labs

44 Including critical values

45 Another patient

46 Clinical decision support: reminders 46

47 Clinical decision support uses allergy information 47

48 Prescribing a medication

49 How about some amoxicllin? 49

50 Oops, patient is allergic 50

51 Maybe erythromycin? 51

52 No, interacts with statin drugs 52

53 Drug interactions for another patient

54 Trying to prescribe nitrates for angina

55 Oops! 55

56 Some more about VistA The pure open-source version is also known as FOIA Vista There are two other streams of VistA activity WorldVistA( follows a more traditional open-source pathway OpenVista ( is more commercially oriented, and some vendors have proprietary extensions from the base code (Medsphere, 2010) There is a small but growing market for VistA, including in other countries (Conn, 2008) 56

57 We also need to think beyond the EHR of a single organization Patients are mobile may develop medical problems or receive care away from their physician office or local hospital Of 3.7M patients in Massachusetts, 31% visited 2 or more hospitals over 5 years (57% of all visits) and 1% visited 5 or more hospitals (10% of all visits) (Bourgeois, 2010) Of 2.8M emergency department (ED) patients in Indiana found 40% of patients had data at multiple institutions, with all 81 EDs sharing patients in common (Finnell, 2011) Also greater need in public health sphere with growing threats of emerging diseases, bioterrorism, etc. 57

58 Beyond the EHR: health information exchange (HIE; Kuperman, 2011) Anytime, anywhere access to clinical information for the care of patients William Yasnoff, MD, PhD Data following the patient Carolyn Clancy, MD, Director, AHRQ Requires that information seamlessly flow across business boundaries Challenges are not only technical, but also financial, legal, etc. But there are other successful examples of information exchange, such as ATM cards, wireless networks, etc. 58

59 Example of HIE: Indiana Health Information Exchange (McDonald, 2005) Access to clinical information in real time by Most hospital emergency departments Many hospital-based clinicians Some primary care providers in community Homeless care network Public school clinics County Health Department Indiana State Health Department 59

60 Results of other HIE efforts have been mixed Successful Inland Northwest Health System (INHS, Spokane, WA Massachusetts ehealth Collaborative ( (Halamka, 2005; Gorroll, 2009) Less so Santa Barbara County Care Data Exchange combination of technical, leadership, and funding problems (Miller, 2007; Brailer, 2007) Portland, Oregon (Conn, 2007) 60

61 Nationwide Health Information Network (NwHIN) HITECH investing $547M in state-level HIE as well as in standards and tools to facilitate NwHIN e.g., Direct Project, wiki.directproject.org 61

62 How much progress have we made? Systematic reviews (Chaudhry, 2006; Goldzweig, 2009; Buntin, 2011) have identified benefits in a variety of areas Although 18-25% of studies come from a small number of health IT leader institutions (Buntin, 2011) 62

63 Caveats about progress HIT may introduce error (Koppel, 2005) or other unintended consequences (Ash, 2004) Report from National Research Council found IT had not met its potential in healthcare (Stead, 2009; good overview in: Conn, 2009 and Conn, 2009) Growing area of concern: HIT system safety (Leviss, 2010; Sittig, 2011) 63

64 Why are we not there? What are the barriers? (Hersh, 2004) Cost Technical challenges Interoperability Privacy and confidentiality Workforce 64

65 Cost barriers Even though there is overall return on investment (ROI), benefit does not accrue to those who pay, especially in small practices (Johnston, 2003) Practices only see 11% of ROI most goes to insurance companies and laboratories But they are usually asked to pay the cost of EHRs Later data showed physicians achieved positive ROI around 2.5 years after initial investment, although range was wide (Miller, 2005) 65

66 Technical challenges While underlying technology (e.g., networks, relational database systems) is well-established, other technical issues remain, such as Implementing systems, especially in office settings (Carter, 2008; Daigrepont, 2011) Matching systems to workflow is essential best systems add time in some areas but make it up in others (Overhage, 2001; Samuels, 2008) Most successful implementations have transformed care delivery and not just replaced paper records (Liang, 2010; Schulte, 2011) 66

67 Need for interoperability Clinical data is trapped in silos, not easily moved from one system to another (Brailer, 2005) Growing push for attention to secondary use of clinical data, which can align benefits for quality assessment, clinical research, public health surveillance, etc. (Safran, 2007) To achieve this, need standards for data elements, communications, etc. (EHRA, 2009; Benson, 2010) 67

68 Concerns about privacy and confidentiality Much written, strong opinions (McGraw, 2009; ACP, 2009) VERY real, but Security technologies are well-known and proven effective Paper-based records are at least as insecure as EHRs and probably more so Human curiosity will trump even best methods, so we need to consider benefits vs. risks HIPAA is a mixed blessing; many argue for modification, e.g., (Ness, 2007; Nass, 2009) 68

69 HIT workforce what do we know? Not much, other than it is important! Case study: implementation of computerized physician order entry (CPOE) showed adverse consequences Mortality rate increased from 2.8% to 6.6% at Children s Hospital of Pittsburgh Pediatric ICU (Han, 2005) Increased mortality not seen at other academic centers (Del Baccaro, 2006; Jacobs, 2006) Pittsburgh adverse outcome may have been avoided with adherence to known best practices (Phibbs, 2005; Sittig, 2006) Problematic health IT implementations well-known, with failure often attributable to lack of understanding of clinical environment (Leviss, 2010) 69

70 Who is the HIT workforce? (Hersh, 2010) Three historical groups of HIT professionals Information technology (IT) usually with computer science or information systems background Health information management (HIM) historical focus on medical records Clinical informatics (CI) often from healthcare backgrounds 70

71 How many HIT personnel do we have and do we need? IT to reach level of known benefit and meaningful use, may need 40,000 (Hersh, 2008) HIM from US Bureau of Labor Statistics occupational employment projections (BLS, 2009) Medical Records and Health Information Technicians (RHITs and coders) about 172,500 employed now, increasing to 207,600 by 2018 (20% growth) CI estimates less clear for this emerging field One physician and nurse in each US hospital (~10,000) (Safran, 2005) About 13,000 in healthcare (Friedman, 2008) and 1,000 in public health (Friedman, 2007) Growing role of CMIO and other CI leaders (Leviss, 2006; Shaffer, 2010) 71

72 What competencies must CI professionals have? (Hersh, 2009) Health and biological sciences: - Medicine, nursing, etc. - Public health -Biology Competencies required in Biomedical and Health Informatics Management and social sciences: - Business administration - Human resources - Organizational behavior Computational and mathematical sciences: - Computer science - Information technology - Statistics 72

73 ONC estimated 51,000 needed for HITECH agenda in 12 workforce roles Mobile Adoption Support Roles Implementation support specialist* Practice workflow and information management redesign specialist* Clinician consultant* Implementation manager* Permanent Staff of Health Care Delivery and Public Health Sites Technical/software support staff* Trainer* Clinician/public health leader Health information management and exchange specialist Health information privacy and security specialist Health Care and Public Health Informaticians Research and development scientist Programmers and software engineer Health IT sub-specialist (to be trained in *community colleges and universities) 73

74 Physicians Other important workforce developments Proposal to establish a clinical informatics subspecialty (Detmer, 2010) based on core curriculum (Gardner, 2009) and training requirements (Safran, 2009) Other health professionals Nursing TIGER initiative (Gugerty, 2009) HIM (Wilhelm, 2007; Dimick, 2008) Nutrition (Hoggle, 2010) 74

75 US has low rates of adoption in inpatient and outpatient settings Adoption in the US is low for both outpatient (Hsiao, 2010) and inpatient settings (Jha, 2010) By most measures, US is a laggard and could learn from other countries (Schoen, 2009) Most other developed countries have undertaken ambitious efforts, e.g., England (Hayes, 2008) Denmark (Protti, 2010) (Hsiao, 2010) NET NZ NOR UK AUS ITA SWE GER FR US CAN (Schoen, 2009)

76 Emerging national consensus is that we need more starts at the top Started with President George W. Bush State of the Union mentioned every year January, 2004 Computerizing health records [can] reduce costs, improve care, and lower the risk of medical mistakes. January, 2007 We need to reduce costs and medical errors with better information technology. Goal of (EHRs) for all Americans by / html Was elevated to even higher priority by President Barack Obama in HITECH Act (ARRA, 2009) 76

77 We are now in a new ARRA of health information technology (HIT) HITECH provides financial incentives for meaningful use of HIT Incentives for EHR adoption by physicians and hospitals (up to $29B) Direct grants administered by federal agencies ($2B) All initiatives overseen by Office of the National Coordinator for Health IT (ONC, 77

78 What is meaningful use (MU) of an EHR? (Stark, 2010; Blumenthal, 2010) Driven by five underlying goals for healthcare system Improving quality, safety and efficiency Engaging patients in their care Increasing coordination of care Improving the health status of the population Ensuring privacy and security Consists of three requirements Use of certified EHR technology in a meaningful manner Utilize certified EHR technology connected for health information exchange (HIE) Use of certified EHR technology to submit information on clinical quality measures 78

79 MU being implemented in three stages HIT-Enabled Health Reform Meaningful Use Criteria HITECH Policies Stage 1 Meaningful Use Criteria (Capture/share data) Stage 2 Meaningful Use Criteria (Advanced care processes with decision support) Stage 3 Meaningful Use Criteria (Improved Outcomes) 79

80 Implementation of MU (Blumenthal, 2010) Implemented through increased Medicare or Medicaid reimbursement over five years to Eligible professionals (EPs) up to $44K Eligible hospitals (EHs) $2-9M There are differences in definitions of above as well as amounts for Medicare vs. Medicaid reimbursement Stage 1 final rules released in July, 2010 by CMS (2010) and ONC (2010) Must achieve core and 5 of 10 menu criteria Summarized in Blumenthal (2010) and many other places 80

81 Stage 1 core criteria (14 for EH; 15 for EP) >30% of unique patients have at least 1 med order entered using CPOE Drug-drug and drug-allergy interaction checks enabled >40% of all permissible prescriptions transmitted electronically (EP only, not EH) >50% of all unique patients have demographics recorded: preferred language, gender, race, ethnicity, dob >80% of all unique patients have at least 1 entry or indication of none on problem list >80% of all unique patients have at least 1 entry or indication of none on medication list >80% of all unique patients have at least 1 entry or indication of none on medication allergy list >50% of patients age 13+ seen have smoking status recorded 81

82 Stage 1 core criteria (cont.) >50% of all unique patients age 2+ have recorded height, weight, blood pressure, calculated BMI, growth charts age 2-20 Implement 1 clinical decision support rule relevant to specialty or high clinical priority with ability to track compliance Report quality measures to CMS provide aggregate numerator, denominator, and exclusions >50% provide patients with an electronic copy of health information upon request within 3 business days Provide clinical summaries to patient for more than 50% of all office visits within 3 business days Performed at least 1 test of certified EHR technology s capacity to electronically exchange key clinical information Conduct or review a security risk analysis and implement security updates as necessary 82

83 Stage 1 menu criteria (require five, one of which must be public health) Implement drug-formulary checks at least 1 internal or external drug formulary for the entire reporting period >50% of all unique patients 65 or older have an indication of an advance directive status recorded >40% of all clinical lab tests ordered are in EHR as structured data Generate lists of patients by specific conditions to use for quality improvement, reduction of disparities, research or outreach Use certified EHR technology to identify patient-specific education resources and provide to the patient if appropriate >50% of transitions of care and referrals by EH provide summary of care record for each transition of care or referral >50% of care transitions perform medication reconciliation Capability to submit electronic syndromic surveillance data to public health agencies and actual submission in accordance with applicable law and practice Capability to submit electronic immunization data to public health agencies Capability to submit electronic laboratory data to public health agencies 83

84 Quality measures differ for EP and EH but required for both EP (outpatient) three required or alternate measures plus three of 13 others, e.g., Hypertension blood pressure measurement Tobacco use assessment and cessation intervention Adult weight screening and follow-up EH (inpatient) 15 required measures, e.g., Diabetes: Hemoglobin A1c, low-density lipoprotein, and blood pressure control Influenza immunization for patients > 50 years old Pneumonia vaccination status for older adults Breast cancer screening Colorectal cancer screening 84

85 MU is just one of several challenges 85

86 Other HITECH funding for the HIT infrastructure Regional Extension Centers (RECs) $677 million for 62 RECs that provide guidance, mainly to small primary care practices and critical access hospitals, in achieving MU State-based health information exchange (HIE) $547 million in grants to states to develop HIE programs Beacon communities $250 million to fund 17 communities that provide exemplary demonstration of MU of EHRs Strategic health information advanced research projects (SHARP) $60 million for four collaborative research centers 86

87 Other funding for the infrastructure: HIT workforce A competent workforce is essential to achieve MU Based on 12 workforce roles, educated in community colleges and universities HITECH funded $118 million for Community college consortia ($70M) Curriculum Development Centers ($10M) Competency testing ($6M) University-based training grants ($32M) 87

88 Community College Consortia to Educate HIT Professionals Program Five regional consortia of 82 community colleges to develop short-term programs to train 10,000 individuals per year in the six community college workforce roles Anticipated enrollment of people with healthcare and/or IT backgrounds probably baccalaureate or higher degrees 88

89 Curriculum Development Centers Program Five universities to collaboratively develop (with community college partners) HIT curricula for 20 components (courses) Oregon Health & Science University (OHSU) Columbia University Johns Hopkins University Duke University University of Alabama Birmingham One of the five centers (OHSU) additionally funded as National Training and Dissemination Center Version 2 of curriculum delivered to community colleges in May, 2011, followed by release to all in July,

90 Components of the ONC HIT curriculum 1. Introduction to Health Care and Public Health in the U.S. 2. The Culture of Health Care 3. Terminology in Health Care and Public Health Settings 4. Introduction to Information and Computer Science 5. History of Health Information Technology in the U.S. 6. Health Management Information Systems 7. Working with Health IT Systems 8. Installation and Maintenance of Health IT Systems 9. Networking and Health Information Exchange 10. Fundamentals of Health Workflow Process Analysis & Redesign 11. Configuring EHRs 12. Quality Improvement 13. Public Health IT 14. Special Topics Course on Vendor-Specific Systems 15. Usability and Human Factors 16. Professionalism/Customer Service in the Health Environment 17. Working in Teams 18. Planning, Management and Leadership for Health IT 19. Introduction to Project Management 20. Training and Instructional Design (Lab components using VA VistA EHR) 90

91 Program of Assistance for University- Based Training (UBT) Funding for education of individuals in workforce roles requiring university-level training at nine universities with existing programs Oregon Health & Science University (OHSU) Columbia University University of Colorado Denver College of Nursing Duke University George Washington University Indiana University Johns Hopkins University University of Minnesota (consortium) Texas State University (consortium) Emphasis on short-term certificate programs delivered via distance learning OHSU program run as tuition assistance program for existing programs 91

92 Telemedicine Delivery of healthcare where time and/or distance separate participants (Field, 2002) Classification of telemedicine (Hersh, 2001) Store-and-forward Office/hospital-based Home-based Ongoing problem is quality of evaluation studies, which impedes coverage by insurers (Hersh, 2006; Ekeland, 2010) Most promising areas may be Home telehealth(darkins, 2008; Shea, 2009) Provider-to-provider communications (Cusack, 2007; McCambridge, 2010) 92

93 Applications of knowledge-based information Information retrieval (IR) systems, aka search engines, are widely available Virtually all of the world s biomedical literature is now available electronically, but significant challenges still exist to its use(hersh, 2008) Evidence-based medicine (EBM) is a proper framework for finding and applying knowledge in clinical care (Straus, 2005; Guyatt, 2008; Guyatt, 2008) Although it must be applied properly in the larger context of healthcare (Haynes, 2002; Cohen, 2004) 93

94 Information retrieval Hersh, 2009 Focuses on indexing and retrieval of knowledgebased information Historically centered on text in documents, but increasingly associated with multimedia and even patient-specific information 94

95 Search systems are popular icons pubmed.gov 95

96 But new problems have emerged 96

97 Overview of an IR system Retrieval Metadata Indexing Queries Content Search engine 97

98 The intellectual tasks of IR Indexing Assigning metadata to content items Can assign Retrieval Subjects (terms) words, phrases from controlled vocabulary, e.g., Medical Subject Headings (MeSH) Attributes e.g., author, source, publication type Most common approaches are Boolean use of AND, OR, NOT Natural language words common to query and content, usually ranked by relevance, e.g., word counts, links (Google), etc. 98

99 Some IR systems have advanced features, e.g., Pubmedlimits, clinical queries, etc.

100 Growing role of IR and related areas in knowledge discovery All literature Possibly relevant literature Definitely relevant literature Actionable knowledge Information retrieval Information extraction, text mining (Hersh, 2009) 100

101 The life-cycle of knowledge-based information many roles for informatics Secondary publications Original research Public data repository Publish Revise Write up results Relinquish copyright Reject Accept Peer review Submit for publication 101

102 Some major IR challenges Open access publishing, especially to taxpayerfunded research (Björk, 2009) NIH Public Access Policy aims to disseminate taxpayerfunded research better (publicaccess.nih.gov) Includes deposit of papers or final submitted manuscripts into PubMed Central (pubmedcentral.gov) Publishers alternative approach is DC Principles for Free Access to Science ( Linkage across silos and other applications, e.g., Digital Object Identifier (DOI) CrossRef project 102

103 IR challenges (cont.) What are best approaches for users to enter queries and for systems to provide output, e.g., Google works well for Web searches, but is it best for biomedicine and health? What about the needs of biomedical researchers, who collect increasing amounts of data and whose research touches on many other areas? What are best approaches for consumers with different levels of education, medical knowledge, etc. (Fox, 2011)? 103

104 Applying EBM Steps of EBM Phrasing a clinical question that is pertinent and answerable Identifying evidence to address the question Critically appraising the evidence to determine if it applies to the patient There are some of the criticisms of the EBM approach (Cohen, 2004) This has led to growing advocacy for Evidence-based practice (Slawson, 2005; Dawes, 2005) Practical clinical trials (Tunis, 2003; Luce, 2009) Comparative effectiveness research (McGinnis, 2007; Murray, 2010) 104

105 Hierarchy of knowledge-based information (evidence) Systems actionable knowledge Synopses concise evidencebased abstractions Syntheses systematic reviews and evidence reports Studies original articles published in journals (Haynes, 2001) 105

106 Where does the best evidence come from? Systems Synopses Syntheses Guidelines, rules, order sets Textbooks, compendia, guidelines Systematic reviews Studies MEDLINE, e.g., Pubmed Journal articles 106

107 Bioinformatics Area of informatics focused on cellular and molecular level Usually associated with genomics, proteomics, metabolomics, and other omics Genome is the collection of all genes in an organism Translational bioinformatics focuses on application in human health and disease (Sarkar, 2011), presaging personalized medicine (Collins, 2010) New areas emerging due to changing nature of bioscience (Gibson, 2009; Gene expression microarrays (Bier, 2008) Genome-wide association studies (GWAS) (Hardy, 2009) 107

108 Informatics challenges become more complex as science advances It used to be so simple the central dogma DNA mrna protein Genes in exons; rest is non-coding (used to be called junk, but is found to be increasingly important) Growing understanding of role of micrornas (Hanrahan, 2011) Increased recognition that genomic variation plays important roles in disease, consisting mainly of Single-nucleotide polymorphisms (SNPs) (Lupski, 2007) Copy-number variations (CNVs) (Redon, 2006; Chen, 2011) 108

109 Informatics challenges (cont.) Key focus now is genetic variation and its impact on health and disease (Hardy, 2009) Constant discovery of new genetic variants associated with development of common and uncommon diseases But need to be careful of spurious associations, e.g., the incidentalome (Kohane, 2006), and apply evidencebased approaches (Attia, 2009 three papers) High-throughput technologies generate much data and information needs, e.g., need for information retrieval and text mining (Cohen, 2005; Krallinger, 2008) 109

110 Increased understanding of molecular basis of disease impacts bioinformatics And vice versa Case in point: cancer Hallmarks of cancer point to cellular activities that reduce cell death while increasing invasiveness (Hanahan, 2011; Johnson, 2011) Growing evidence that these changes enabled by changes in genome and its function (Cowin, 2010) These changes are manifested temporally (Durinck, 2011) and more readily measured with newer sequencing technologies (Meyerson, 2010) New resources, such as the Cancer Genome Atlas (Chin, 2008) and PharmGKB(Thorn, 2010), will guide research, diagnosis, and treatment 110

111 Melding of bioinformatics and clinical informatics: emergenetwork Consortium aiming to link growing number of biorepositorieswith sequenced DNA to data in EHR systems for large-scale, high-throughput genetic research (McCarty, 2011; Wilke, 2011) Map the phenotype (EHR and tissue) to the genotype (DNA sequences) using informatics Some early work includes use of NLP for Replicating finding of known gene-disease associations from research data in EHR data (Denny, 2010) Discovering new associations (Denny, 2010) Mapping to the phenotype, including controlled terminologies (Pathak, 2011) 111

112 Informatics education and training Since a highly multi-disciplinary field, no standard curriculum or accreditation Listing of US programs on Web site of American Medical Informatics Association Description of OHSU program to follow as an example; consult other programs Web sites for details on their programs International perspective from International Medical Informatics Association (IMIA) recommendations (Mantas, 2010) Education has historically focused on academics but is evolving to meet the needs of practitioners and users 112

113 Career pathways have diverse inputs and outputs Health care professions, e.g., medicine, nursing, etc. Natural and life sciences, e.g., biology, genetics, etc. Computer science (CS), IT, and undergraduate informatics Health information management (HIM) Graduatelevel biomedical informatics education Jobs in: Health care systems Biomedical research Industry Academia Others Others, e.g., business, library and information science 113

114 Biomedical informatics education at OHSU All at graduate level Academic Predoc/Postdoc Fellowship funded by NLM and VA PhD in Biomedical Informatics degree Master of Science in Biomedical Informatics degree for postdocs from other fields Professional Master of Science and Master of Biomedical Informatics degrees Graduate Certificate Program (distance learning) Liaison OHSU-AMIA 10x10 program 114

115 Informatics curriculum at OHSU general principles Aims to cover the full spectrum of biomedical and health informatics (Hersh, 2005; Hersh, 2007) Curriculum centered around knowledge base Core knowledge at master s level PhD adds advanced courses and research Building block approach allows progression to higher levels Have established three tracks Clinical informatics Bioinformatics and computational biology Health information management Aiming to establish others, e.g., public health informatics 115

116 Knowledge base and its domains Biomedical Informatics Health Care Bioinformatics and Computational Biology Biomedical Sciences Organizational and management sciences Computer Science Evaluative Sciences Biostatistics Electives and graduation requirements Clinical Informatics Track Bioinformatics Track 116

117 Building block approach to curriculum Graduate Certificate -Tracks: - Clinical Informatics - Health Information Management 10x10 - Or introductory course Masters -Tracks: - Clinical Informatics - Bioinformatics - Thesis or Capstone PhD - Knowledge Base - Advanced Research Methods - Biostatistics -Cognate - Advanced Topics - Doctoral Symposium - Mentored Teaching - Dissertation 117

118 Educating the liaisons 10x10 Partnership with American Medical Informatics Association (AMIA) to meet Safran s(2005) stated need to educate one physician and one nurse from each of the 6000 US hospitals in informatics Original goal to educate 10,000 healthcare professionals by 2010 Course consists of introductory on-line course and adding one-day face-to-face session Initial offerings well-received (Hersh, 2007; Feldman, 2008) Over 1000 have completed by 2010 About 15% going on to further study Program has continued beyond 2010 Other partners are also offering courses 118

119 Topics of OHSU 10x10 course Overview of Field and Problems Motivating It Biomedical Computing Electronic and Personal Health Records (EHR, PHR) Standards and Interoperability; Privacy, Confidentiality, and Security Meaningful Use of the EHR EHR Implementation and Evaluation Evidence-Based Medicine and Medical Decision-Making Information Retrieval and Digital Libraries Imaging Informatics and Telemedicine Translational Bioinformatics and Personalized Medicine Organizational and Management Issues in Informatics 119

120 Educating beyond OHSU distance learning (Hersh, 2001) Initially in Graduate Certificate, later master s Teaching modalities include Voice-over-Powerpoint lectures Threaded discussions Readings, virtual projects, etc. Courses are not correspondence courses; interaction is a core component Have created a virtual community Meet at AMIA, HIMSS, OHSU, etc. 120

121 New models for education can be developed with this technology Delivery to many parts of the globe, e.g., South America, Africa, etc. Translation of 10x10 course into Spanish for Latin American audience (Otero, 2010) Offered in partnership with Hospital Italiano of Buenos Aires, Argentina Over 600 participants from across Latin America have completed course 121

122 Current and future directions OHSU program continues to innovate and grow Over 600 have matriculated since inception Over 300 alumni with jobs in healthcare settings, academia, industry, and elsewhere Funding for study of clinical informatics available for eligible students through, among others, ONC UBT program winding down with end of HITECH funding National Library of Medicine Training Grant 122

123 Bringing it all together To achieve the vision of BMHI, we cannot lose focus of the information as well as the scientific processes to determine how to use it most effectively Informatics-trained professionals will lead the use of IT to improve healthcare and biomedical research Exciting research areas on the horizon Improving healthcare delivery Enabling of patients and consumers in healthcare Genomics, bioinformatics, and personalized medicine Mining data to analyze and generate hypotheses for future biomedical research And how will the grand experiment of HITECH play out? 123

124 For more information Bill Hersh Informatics Professor blog OHSU Department of Medical Informatics & Clinical Epidemiology What is BMHI? Office of the National Coordinator for Health IT American Medical Informatics Association 124