Tag Archives: CPOE

Step 1 – Opposing Views of EMR’s Ability to Improve Care and a Possible Synthesis

The Argument In Favor of: Computerized electronic medical records (EMR) will improve quality of care. EMR facilitates streamlining administrative processes, reducing overhead. Accurate and quickly accessible patient health information is a prerequisite to timely, informed, patient-centered medical care. Numerous studies have shown that CPOE can reduce medication errors and adverse events as much as 99%, increasing safety and reducing costs.(1) The ability for practitioners to access the same record in real-time from multiple sites or to send a record electronically to another provider puts potentially life-saving information where it is needed most. Decision support systems built on top of EMRs can support care by managing clinical complexity, controlling cost by suggesting less expensive alternatives, catching drug-drug or drug-allergy interactions, and promoting best practices.(2) EMR can help empower patients by connecting them to tailored health education materials. Other information intensive industries spend approximately 10% of their budgets on IT whereas health spends only 3%. If the health sector spent similarly, it would be able to realize significant gains.

The Argument Against: Electronic medical records rarely improve medical care and can even make it worse. Jeffrey Linder et al found that there was no association with presence of EMR and quality for 17 different measures, and this has been confirmed by other studies.(3) Providers who have experienced gains are generally academic medical centers whose results are not reproducible outside of that setting. In one example, Children’s Hospital of Pittsburgh rolled back a multi-million dollar CPOE implementation in the pediatric ICU after it was discovered that mortality had increased. Physician productivity can drop as much as 20% for the first 6 months after EMR implementation. A good ROI has generally only been obtained by large, integrated networks through savings on administrative overhead. Until technology systems mature and implementation processes improved, resources would be better invested elsewhere.

A Potential Synthesis: Electronic medical records are an enabling technology that supports cost-savings and quality improvement processes only if meaningfully and effectively used. David Cutler maintains that other industries required ten years to realize industry-wide gains from the use of information technology. The health sector started using IT later than other industries, but will be able to realize significant gains after clinical workflows and local cultures adapt. EMR data enables providers to do monitoring and evaluation and quality improvement that would not be possible otherwise, but business processes must be modified to take advantage of them. Providers should first implement technologies and features that have proven to be effective, such as CPOE, automated prescribing and dispensation. National “meaningful use” regulations, while imperfect and politicized, help guide physicians, health system planners and vendors on methods to increasingly leverage technology to improve health.

—-

1. Koppel, JAMA 2005; Bates, 1998; Pestotnik 1996.
2. Perreault L, Metzger J. A pragmatic framework for understanding clinical decision support. Journal of Healthcare Information Management. 1999;13(2):5-21.
3. Jeffrey A. Linder, MD, MPH; Jun Ma, MD, RD, PhD; David W. Bates, MD, MSc; Blackford Middleton, MD, MPH, MSc; Randall S. Stafford, MD, PhD. Electronic Health Record Use and the Quality of Ambulatory Care in the United States. Arch Intern Med. 2007;167(13):1400-1405.

Health IT Ontology

2 weeks ago I solicited help to put together this Health IT Ontology (see Components of HIT…a start). This post is the result of 6 rounds of edits. The new name, I think, better represents the goal of defining all the entities and relationships within the domain of health IT. Click on the image below to see it full size.

Health IT Ontology

Following are the top-level categories:

  • Health Information Technology
    • Clinical Information System
    • Hospital/Clinic Management
    • Consumer-Oriented Technologies
    • Public Health & Biosurveillance
    • Reference
    • Research
    • Regional & System-Level HIS

The initial motivation behind this was to determine where ART-focused EMRs sat in the scope of HIT, but what I expected to be a trivial exercise quickly became a difficult task. Health IT is an extremely complex and expansive domain and every item in this heirarchy could be broken down into even smaller pieces (similar to EMR/EHR). My goal for this diagram was to cover the breadth of health IT more than the depth. It is certainly possible that there are some oversights, in which case I would love to hear from you.

I welcome your thoughts, criticisms, and suggestions on the HIT Onthology. Using social media (esp. Twitter and Aardvark) was so successful this time around that I plan to pursue more online collaborative projects in the future.

Many thanks to everyone who contributed, and a special shout out to Jacob Sattelmair, Janette Heung, blog commenters, Richard Thall and Eddie from Aardvark, and the score of Twitterers who provided very valuable feedback!

Components of HIT…a start

UPDATE: The final output of this project, HIT Ontology, can be found in this blog post: https://singularityblog.wordpress.com/2009/07/20/health-it-ontology/

Health information technology (HIT) is a broad and extremely complex field, and I want to visualize it. I’m going to need your help to do it. But first it needs defining…

HIT could simply be defined as any information technology utilized within the healthcare industry vertical, but that would be too inclusive, because that means a MySQL database is considered HIT because it is sometimes used in a hospital. Brailer & Thompson, former ONC Secretary and former HHS Secretary respectively, define it as “the application of information processing involving both computer hardware and software that deals with the storage, retrieval, sharing, and use of health care information, data, and knowledge for communication and decision making” (Thompson & Braile, 2004). The line between HIT and health informatics is fuzzy and we’ll ignore it for now.

With this definition, I tried to create a hierarchical list of the types of health IT software. I want the list to be comprehensive in breadth and don’t care quite as much about depth (3 or 4 levels should be sufficient). There are dozens of ways to structure this list and probably hundreds of items I missed. This is a work in progress, so please leave a comment and let me know what you would change/add/remove. I’ll keep updating it until everyone feels good about it. After that comes the visualization…

HIT Categorization Hierarchy – Take 5

  • Clinical
    • EMR/EHR
      • Ambulatory
      • Specialty
      • Anti-Retroviral Treatment (ART) Focused (common in areas with high HIV/AIDS & TB prevalence)
    • eRx (CPOE)
    • Clinical Decision Support
    • Digital Imaging & Archiving Systems (e.g. PACS)
    • Medical Devices & Equipment
    • Clinical Document Management
    • “Personalized Medicine”
  • Hospital/Clinic Management
    • Physician Office Management Information System (POMIS)
    • Hospital Management Information System (HMIS)
    • Accounting
    • Patient Billing
    • Claims Processing
    • Human Resource Management
    • OR Scheduling
    • Appointment Scheduling
    • Lab/Pharmacy Management
  • Public Health & Biosurveillance
    • Public Health Reporting
    • Diesease Surveillance Networks (e.g. CDC Biomonitoring and Environmental Public Health Tracking Network)
    • Vital Registry (Birth, Death, & Marraige Records)
  • Consumer-Oriented Technologies
    • Personal Health Devices (e.g. WAN-enabled weight scale, phone-enabled glucose monitor, etc.)
    • Personal Health Applications (i.e. exercise & weight tracking)
    • Patient Portals
    • Personal Health Records (PHR)
    • Health-centered Social Networks (Patients Like Me, 23andme, etc.)
  • Medical References
    • Drug references (for docs and patients)
    • Medical references (like WebMD, also for docs and patients)
  • Research
    • Genomics
    • Medical data warehousing
    • Clinical Trial Recruitment, Management, etc.
  • Regional & Systems Level Health Information Systems
    • Vitals Registration
    • Health Information Exchange (HIE)
    • National Health Information Network (NHIN)

A special thanks to the Twitterers that have already helped me on this: @chadosgood, @oneofthefreds, @ChristineKraft, @ePatientDave, @MedC2, and my good friend Jake. And a shout out to Sam Adam’s HIT Primer on his blog, IT (R)EVOLUTION, that helped get me started.

A few other helpful sources:

US Behind in HIT Spending – Stimulus Insufficient

Despite the fact that the US spends nearly twice as much on healthcare as any other country, the US is as much as 12 years behind other OECD countries in health information technology investment. See the Commonwealth Fund’s entry on Health Care Spending and Use of Information Technology in OECD Countries.

hit-efforts-in-six-countries

The American Recovery & Reinvestment Act of 2009–the Stimulus Package–apportions $19 billion for investment into the HIT infrastructure in the US. As much as $3 billion goes to the Office of the National Coordinator (which will now be codified) and other standards creating bodies. The remaining amount will be given to providers primarily through increased Medicare reimbursement. If divided evenly, each hospital would receive approximately $11 million. A substantial sum, but hardly close to the $200 million over 3 years required in a typical implementation at a 300+ bed hospital. Only 10% of hospitals currently have full electronic health records. Another 20-30% are in planning or implementation stages. The stimulus may encourage more providers to enter the planning stages and will help along those already in the process during difficult economic times. But $11 million for the remaining 60-70% is entirely insufficient.

Evidence shows that the only providers that stand to get a return on investment in HIT are large network providers with geographically distributed practices, such as Kaiser or the VA. This makes sense, as the administrative cost of sharing information is high. The early adopters (the 10%) consist of these large networks and a few providers with well-funded, forward-thinking CIOs. The 20-30% currently planning hope to break even at best and justify the investment by improved patient care (especially through CPOE). The rest are mostly too small to realize significant cost savings and will likely need much more than $11 million to break even.