The federal government announced regulations this evening that define “meaningful use” of EHRs and the CMS incentive program associated with it, barely meeting the December 31 required date for issuing an initial set of standards.
The rules will go into effect 30 days after publication following a public comment period. The meaningful use rule is here (warning: PDF).
The incentive rule (all 556 pages of it) is here (warning: PDF). It contains specifics about percentages of orders, payment schedules, specific numerators and denominators for measures, etc. I gave it a quick skim and got most of the information about use measures, but if someone wants to summarize the payment portion early Thursday, I will post it (since I’ll be at work).
These specifications apply to Stage 1, which take effect in 2011. They fall into four categories of standards: vocabulary, content exchange, transport, and privacy and security.
Stage 2 requirements start in 2013 and Stage 3 requirements in 2015. Those will be defined later by HHS.
This is a summary of the most important information.
CPOE
Practices: Use CPOE for orders involving medications, laboratory, radiology, and referrals.
Hospitals: medications, laboratory, radiology, blood bank, PT, OT, RT, rehab, dialysis, consults, and discharge and transfer.
Orders do not have to be sent electronically to the fulfilling department (lab, pharmacy, etc.)
Practices must enter 80% of their total orders directly by the clinician into the CPOE system. Hospitals must have 10% of all orders entered by CPOE.
Clinical Checking of Orders
Real-time screening (drug-drug interactions and drug-allergy contraindications), formulary check, user ability to maintain screening rules, track user responses to alerts.
Problem List
Longitudinal current and active diagnoses coded in ICD-9-CM or SNOMED CT.
80% of unique patients must have at least one coded problem/diagnosis, with “none” being an allowed entry (hospitals and practices).
E-Prescribing
Practices only.
Must send 75% of non-controlled substance prescriptions electronically.
Active Medication List
80% of unique patients must have at least one coded entry, with “none” being an allowed entry (hospitals and practices).
Medication Allergy List
Longitudinal with allergy history.
80% of unique patients must have at least one coded entry, with “none” being an allowed entry (hospitals and practices).
Demographics
Practices: preferred language, insurance type, gender, race, ethnicity, and data of birth.
Hospitals: all of the above plus date and cause of death if applicable.
80% of patients must have demographics recorded as structured data
Vital Signs
Height, weight, BP, BMI, growth charts for patients 2-20 years old, temperature, pulse.
80% of patients aged 2 and over must have blood pressure and BMI entered.
Children 2-20 must have a growth chart.
Smoking Status
Record if current smoker, former smoker, or never smoked.
Must be recorded for 80% of patients.
Structured lab results
Display results, translate LOINC codes, allow maintenance based on new results.
Must record as structured EHR data 50% of all results that are delivered in positive/negative or numeric format.
Patient Lists
Allow user to select, sort, retrieve, and output patient lists based on demographics, medications, and conditions.
Report Quality Measures to CMS and States
Calculate, display, and submit quality measure results
Patient Reminders
Practices only: issue based on patient preferences, demographics, conditions, and medication list.
Five Clinical Decision Support Rules
Beyond drug screening, based on demographics: diagnoses, lab results, or medication list. Real-time alerts and suggestions based on evidence. Track response to alerts.
Eligibility
Allow user to record and display based on eligibility response from insurer.
Must cover 80% of unique patients.
Submit Claims
Must submit 80% of all claims filed electronically.
Electronic Copy of Health Information to Patients
Allow user to create an electronic copy of test results, problem list, medication list, medication allergy list, immunizations, and procedures. Hospitals must also provide a discharge summary but not procedures.
Must provide an electronic copy of health information to requesting patients within 48 hours.
Electronic Copy of Discharge Instructions
Hospitals only.
Must provide electronically to 80% of discharged patients who request them.
Timely Patient Access to Health Information
Practices only: diagnostic results, problem list, medication list, medication allergy list, immunizations, and procedures. Within 96 hours of availability.
Must provide to 10% of unique patients.
Clinical Summary of Each Office Visit
Practices only: diagnostic results, medication list, procedures, problem list, immunizations.
Must provide for 80% of office visits.
Information Exchange
Enable electronic sending and receiving of diagnostic test results, problem list, medication list, medication allergy list, immunizations, and procedures. Hospital requirements also include a discharge summary.
Must conduct at least one test of exchanging information.
Medication Reconciliation
Compare and merge two or more medication lists into a single list that can be displayed in real time.
Must be performed in 80% of encounters and care transitions.
Submit Data to Immunization Registries
Must conduct at least one test of submitting information.
Submit Lab Results to Public Health Agencies
Hospitals only.
Must conduct at least one test of submitting information.
Submit Syndrome Surveillance Data to Public Health Agencies
Must conduct at least one test of submitting information.
Protect Electronic Patient Information
Unique identifier, emergency access for authorized users, session timeout, encryption where preferred, encryption when exchanging information, maintain audit logs, provide integrity check for recipient of electronically transmitted information, verify user identities and access privileges, record PHI disclosures.
Must conduct a security risk analysis and implement security updates.
Transport Standards
SOAP and REST
HL7 CDA R2 Level 2 CCD or ASTM CCR
ICD-9-CM or SNOMED CT for problem lists
ICD-9-CM or CPT-4 for procedures, moving to ICD-10-PCS or CPT-4 for Stage 2
RXNorm for medication lists
UNII for Stage 2 allergy lists (no standard now)
CDA template for Stage 2 vital signs (no standard now)
UCUM for Stage 2 units of measure (no standard now)
LOINC for lab results
NCPDP Formulary & Benefits Standard 1.0 for drug formulary checks
NCPDP SCRIPT 8.1 or 10.6 for prescription information
ASC X12N and NCPDP for transactions
CMS PQRI 2008 Registry XML for quality measures
HL7 2.5.1 for submitting lab results to public health agencies, with UCUM and SNOMED CT encouraged
HL7 2.3.1 or 2.5.1 for submitted surveillance information to public health agencies and for immunization information
Encryption only if organization sets it as a standard
Median Estimated One-Time Costs for CCHIT-Certified EHRS to Be Certified as Complete EHRs
CCHIT Ambulatory 2008: $1 million
CCHIT 2007/2008 Inpatient: $1.38 million
Median Estimated One-Time Costs for Pre-2008 or Uncertified EHRS to Be Certified as Complete EHRs
Practice EHR: $2.4 million
Hospital EHR: $3.3 million
Estimated Median Industry Costs for EHR Preparation
2010: $61.35 million
2011: $54.53 million
2012: $20.45 million
From C’mon Man: “Re: would you buy an EHR from this man? Or a demonstration of how easy it is to smile at the patient, hold the computer, and enter data all at the same time. I do not get it, why is anyone fussing? This ad has sold me, outdated CCHIT and all.” Hey, have some holiday compassion: it’s tough making a living trying to get doctors to use EMRs they don’t really want. My first thought reading the “gift that keeps on giving” part of the ad: the old joke about syphilis.
Note to Mississippi Governor Haley Barbour: don’t ask a question if you don’t want to know the answer. The Gov, getting his tweet on, sends out a blurb pitching cost cutting. An administrative assistant in University Medical Center’s nursing school tweets back, suggesting that maybe he should get his medical exams during normal working hours like everybody else instead of requiring employees to come in after hours on overtime. The Governor’s Office is not appreciative, tracking her down and demanding that the hospital’s compliance officer deal with her. They did, citing HIPAA laws in telling her to quit or be fired even though she didn’t know anything about his health first-hand. The Governor’s Office claims they didn’t contact anyone.
I just noticed that the verified e-mail subscriber count has passed 5,000. Thanks to everybody who reads HIStalk. I can’t express how satisfying and humbling that is, especially when I’ve had a sucky day at work (not today, though – it’s great with everyone taking time off, although the long winter grind starts in earnest next week).
From Thanks: “Re: KLAS. Thank you for publishing the article on KLAS. I was really upset that you never said much lately about this. KLAS is a big scam.” The Readers Write article by Swearingen Software CEO Randall Swearingen drew quite a few diverse comments. Some believe KLAS is an evil money factory, while others say their approach is reasonable. Not that you care, but here are my observations about KLAS.
- I have contributed to KLAS surveys (although not recently) and never detected any suggestion of impropriety. I found their information useful and referred to it fairly often, although not to the exclusion of doing my own homework. I wouldn’t have paid for the subscription and reports.
- I would like to see more statistical transparency in their methods, preferably by external and impartial oversight. Adam Gale said he welcomed this in my 2007 interview with him, but I haven’t seen any changes.
- I don’t believe it when KLAS insists that wild result swings (the “first-to-worst” phenomenon) is a reflection of vendor changes. I think it highlights the problem of trying to extrapolate hard statistics from squishy interview data, no matter how many mumbo-jumbo graphs you include.
- KLAS doesn’t claim to be the Consumer Reports of the industry (see Adam’s comments in my interview). They are a survey company, not a software testing company. At best, they accurately summarize information that vendor customers have given them.
- KLAS has always taken specific data of limited usefulness and wildly extended it into all kinds of repurposed reports that mean very little but that provide extra sales revenue. I have always ignored those anyway, so I can’t say that bothered me.
- The KLAS business model is the same as that of HIMSS: providers pay little to nothing, but their participation motivates vendors to pay to play. Whatever they are selling, vendors keep buying of their own free will.
- Like every other survey-based award, vendors who score well plaster their results everywhere. Those who don’t complain that the process was rigged.
- For me, I paid the most attention to the user comments rather than the fancy graphs and stoplights. For we provider-siders, I bet I could provide an equally valuable service by just contacting a lot of verified system users, asking them a handful of questions, and publishing the results.
- My overall conclusion: the evils of KLAS are really a reflection of the evils of its provider and vendor members. Vendors try to game the system without getting caught, while providers unwisely overweight the value of KLAS in making their IT decisions. All of that is highly profitable to KLAS, but more power to them for creating a niche that still has minimal competition and strong business after all these years.
Back in 2005, I wrote an editorial pitching the idea of a standard healthcare database schema. I’ve seen other folks pick up that idea lately. Given the push for interoperability, I still like the idea. Here’s a snip of what I said then:
This is where my noodling got out of hand. Why can’t every vendor voluntarily or mandatorily use the same database layout for core information? How many ways can you express and repose standard elements such as date of birth, gender, address, etc.? Vendors can, when under duress, feed their data to a standard interface. Why can’t all systems just use an approved core set of tables, updated by the same core set of business rules, and then add their value through additional related tables, GUIs, business rules, etc.? Everyone’s patient database could look and work the same. Seen one, seen ‘em all. Customers would be as thrilled by this idea as vendors would be appalled by it. Standard reports would work for every hospital, not just those of a particular vendor. Data translation for third-party reporting would be a no-brainer. Conversion of one system to another would be a piece of cake. Hospitals could easily merge and un-merge with each other to their heart’s content, with data conversion and extraction being assured. You might even have your choice of database software, given an Internet-like abstraction layer that supports everything from Oracle to Cache’. Talk about your interoperability!
An unconvincing article a couple of months ago concluded that remote monitoring of ICU patients by intensivists had little impact on outcomes. I can’t see the full text of this new JAMA article, but it seems much more conclusive, even though its conclusion is the same: “Remote monitoring of ICU patients was not associated with an overall improvement in mortality or LOS.” It’s the CPOE problem, however – many of the institutions had it, but weren’t really using it (although that in itself might, as for CPOE, give an organization reason to question its own capabilities before whipping out the checkbook).
LifeBot announces GA of its VoIP-based workstation for EMS telemedicine, offering full compatibility with digital radio systems.
Inga’s got a couple of great interviews running on HIStalk Practice: Scott Decker (new president of NextGen) and William Zurhellen, MD (a pediatrician and CCHIT expert panel member who has some shockingly frank things to say about the state of EMRs, CCHIT, and standardization).
Listening: Ben’s Brother, slightly whiny Britpop that still sounds good, although I eventually needed some nasty chick music to offset it and headed over to desk-drum to L7 for the zillionth time.
OHCHIT has an upcoming conference call to talk about the $6 million it will spend to get universities to develop a health IT competency exam (warning: PDF) for degree-less HIT people, a little chunk of its $120 million Health IT Workforce Development Program.
Greenway Medical rolls out its PrimeSuite EHR, PM, and interoperability product to Bethesda Healthcare System (FL).
Northwestern Medical Center (VT) gets CON approval to implement Meditech for $5 million, also expecting $577K in stimulus money as a result.
Odd lawsuit: a man sues Barnes Jewish Hospital after he claims he slipped on a Q-tip while visiting a patient, causing extensive injury, disability, and suffering.
E-mail me.
Submit your article of up to 500 words in length, subject to editing for clarity and brevity (please note: I run only original articles that have not appeared on any Web site or in any publication and I can’t use anything that looks like a commercial pitch). I’ll use a phony name for you unless you tell me otherwise. Thanks for sharing!
TPD’s Review of the RIS/PACS Relationship
By The PACS Designer
As hospitals try to get more efficient, it would be a good time to review what the Radiology Information System (RIS) and Picture Archiving and Communications System (PACS) can bring to the institution when it comes to efficiency.
First, let’s review the imaging piece, which is PACS. The main purpose of the PACS is to digitize image files for easier access and increased image sharing. While a PACS is a significant change, it does start to improve processes through more rapid access to image files. The PACS also encourages the sharing of image information with other departments.
Next, the acquisition of a PACS can be a significant draw on financial resources, as it will require workstations for each radiology department member, and other need-to-know individuals who require image file access privileges.
Weighing the cost against the benefits of a PACS, the institution can reduce film and chemical costs with a PACS, and also improve process flow for patients through quicker access to image files. These improved results have to be weighed against the financial outlay that has to be made to bring digital imaging to Radiology.
Now, adding a RIS to a PACS can further improve the scheduling of patients for Radiology procedures. The RIS allows efficient scheduling to take place through its automating of the scheduling software. The software can also highlight potential bottlenecks to alert staff to a looming problem.
The RIS lets everyone know what each radiologists workload is, and how fast equipment can be used to take advantage of each equipments efficiency features.
Another benefit of adding a RIS to a PACS is the bi-directional flow of patient information after procedures are completed and sent back to the RIS for staff review and planning.
In summary, a RIS/PACS configuration can bring great value to the Radiology imaging process, and help reduce costs overall after careful redesign of existing processes.
Awards For Sale?
By Randall Swearingen
KLAS recently named its “Top 20 Best in KLAS Awards: Software & Professional Services 2009” report. Before I list my concerns, you need to understand a few basic points about KLAS.
Their main award is the “Best in KLAS” award. It is supposed to be awarded to the vendor with the highest customer satisfaction scores in a given category (i.e. the best vendor). To be “Best in KLAS”, there has to be a minimum of three non-asterisked vendors in a given category. Vendors are asterisked when they have less than 15 customer surveys because KLAS doesn’t consider the data reliable. In addition to their “Best in KLAS” award, KLAS also has “Segment Leader” awards for those vendors whose categories don’t qualify for “Best in KLAS”.
It is important to note where KLAS gets their revenue because it seems to indicate a conflict of interest between serving the healthcare industry and serving the healthcare vendors. One source of their revenue is from selling their reports to hospitals, clinics, consultants, vendors, etc. Since hospitals and clinics can get free reports by completing just one vendor survey, very little revenue comes from them. The bulk of KLAS revenue comes from vendors. Vendors pay KLAS to survey enough of their customers to get the asterisks removed their products. That isn’t cheap.
KLAS further encourages vendors to pay to have the asterisks removed from at least two of their inferior competitors so that they can be eligible for “Best in KLAS”.
KLAS also charges vendors an annual fee to view KLAS data (including their own). The fee is calculated as a percentage of that vendor’s annual revenue. Thus, larger companies pay more than smaller companies to view KLAS data.
Of course most vendors elect not to pay KLAS, which is why most products are asterisked in their database. But, those who do pay and who are awarded “Best in KLAS” play the award up big time in ads, trade shows, etc.
See the conflict of interest yet? Isn’t the purpose of KLAS to identify and reward the best vendors on the basis of customer satisfaction? Not based on how much a vendor pays?
Back to this year’s report. As a radiology information system vendor, I went straight from the e-blast to review the radiology winners. The “Segment Leader” in the Radiology Ambulatory category this year went to a vendor who happens to be asterisked. Upon reviewing the report, I contacted one of my customers, who has a KLAS account, and asked them to compile some KLAS data for me. Turns out that the winning vendor had scores that were slightly better than those of Swearingen Software.
I then turned my focus to the Radiology Small category. Swearingen Software had the highest scores in the Radiology Small category in all three sections (PRIMARY INDICATORS, DETAIL INDICATORS, and BUSINESS INDICATORS) but the “Segment Leader” award was given to a vendor whose scores ranked seventh out of the10 vendors in all three sections! If you have a KLAS account, you can easily verify all of this information. In the KLAS e-blast, they did not disclose how the “Segment Leaders” were selected or that it doesn’t necessarily go to the vendor with the highest scores.
I felt compelled to dig deeper, so I asked my customer to review the “Segment Leader” section of the report and look for any clues that might explain this action. My customer informed me that upon close inspection of the Top 20 KLAS report on their Web site, a small note is shown below the “Segment Leader” chart which states: “Other solutions must have at least two products that meet the KLAS minimum for statistical confidence in order for a product to earn category leader status.” (That means having a minimum of two non-asterisked products somewhere in KLAS).
OK. So let me get this straight. It’s possible for a vendor to have two non-asterisked products (even if they are the absolute worst scores in their respective categories) AND they can have the absolute worst score in a different category AND they can still win the “Segment Leader” award for that category. Remember, vendors have to pay to get their asterisks removed. Hmmm. What happened to the concept of the award going to the vendor with the best scores?
Simple questions: who monitors KLAS? Who audits them? What independent source verifies their data to make sure it is accurate and fairly represented since they seem to have influence over some buying decisions? Answer: nobody.
I think the “Best in KLAS” award should be renamed to the “Deep Pockets” award. It would be more fitting.
Randall Swearingen is founder and CEO of Swearingen Software, Inc. of Houston, TX.
Medical imaging vendor AMICAS announced this morning that it will be bought by private equity firm Thoma Bravo LLC for $5.35 per share, a 21% premium to Thursday’s closing price.
Stephen Kahane, AMICAS president, CEO, and chairman, was quoted as saying, “With the additional capital and operational expertise available to AMICAS through Thoma Bravo, we will be able to grow as the needs of our customers evolve and will be enabled to better serve our market.”
