Sunday, December 30, 2012

Public Comment by Feb 4: ONC Health IT Safety Plan

ONC recently released its Health IT Patient Safety Action and Surveillance Plan for public comments, which are due by Feb 4, 2013. Comments can be sent to

The report can be found HERE, as well as a brief Fact Sheet.  The report is related to the IOM's 2011 recommendations from Health IT & Patient Safety: Building Safer Systems for Better Care.


Patient safety objectives

  1. Use health IT to make care safer 
  2. Continuously improve the safety of health IT

Increase the quantity and quality of data and knowledge about health IT safety

  1. Make it easier for clinicians to report patient safety events and risks using EHR technology.
  2. Engage health IT developers to embrace their shared responsibility for patient safety and promote reporting of patient safety events and risks. 
  3. Provide support to Patient Safety Organizations (PSOs) to identify, aggregate, and analyze health IT safety event and hazard reports.
  4. Incorporate health IT safety in post-market surveillance of certified EHR technology through ONC-Authorized Certification Bodies (ONC-ACBs).
  5. Align CMS health and safety standards with the safety of health IT, and train surveyors. 
  6. Collect data on health IT safety events through the Quality & Safety Review System (QSRS).
  7. Monitor health IT adverse event reports to the Manufacturer and User Facility Device Experience (MAUDE) database.

Target resources and corrective actions to improve health IT safety and patient safety

  1. Use Meaningful Use of EHR technology to improve patient safety.
  2. Incorporate safety into certification criteria for health IT products.
  3. Support research and development of testing, user tools, and best practices related to health IT safety and its safe use.
  4. Incorporate health IT safety into medical education and training for all health care providers.
  5. Investigate and take corrective action, when necessary, to address serious adverse events or unsafe conditions involving EHR technology. 

Promote a culture of safety related to health IT

  1. Develop health IT safety priority areas, measures, and targets.
  2. Publish a report on a strategy and recommendations for an appropriate, risk-based regulatory framework for health IT. 
  3. Establish an ONC Safety Program to coordinate the implementation the Health IT Safety Plan. 
  4. Encourage state governments to incorporate health IT into their patient safety oversight programs.  
  5. Encourage private sector leadership and shared responsibility for health IT patient safety.

Saturday, December 1, 2012

HIMSS mHealth Summit 2012 in DC

I'm looking forward to the mHealth Summit in DC on Monday, Dec 3. I've not been to this conference before, but mobile health apps and and the use of mobile devices in the service of health care delivery and education has become such a game changer that one ignores this space with great peril.

A quick search of the site for the terms behavioral or mental or psychiatry or psychology gets 75 hits! While a lot of these terms are in the bios for speakers and such, this is still quite a bit when compared to other medical specialty terms, such as cardiology (6), surgery (5), neurology (2), dermatology (3), and radiology (5).  Granted, psychiatry only gets you 4 hits compared with 23 for psychology and 50 for behavioral, but considering all of these terms provides a lot more hits compared to other specialty areas. Even the term mental appears in 24 hits. Why is that that behavioral health is so represented?

I expect it is because the speakers and topics in the mhealth area will reflect three major components of healthcare:

  • patients currently coalesce around disease interests, especially those conditions that lend themselves to peer support
  • providers who treat or speak about these disease interests will naturally follow
  • healthcare costs: because mhealth is likely to be most disruptive in the disease areas that are most costly, it makes sense that these areas will be more heavily represented at this conference. Example: 2011 Maryland Medicaid data shows that people with chronic medical problems are admitted 8-15 times more often if they have comorbid mental health and substance abuse diagnoses. 8-15 times!! (data available on request)
AHRQ listed the top 10 most expensive conditions (2008 data). Here they are with the mean annual expenditures (averaged for men & women combined) and the number of hits on the website.
  1. Cancer $4678... 515 hits (cancer|oncology)
  2. Heart disease $4043... 510 hits (heart|cardiac|cardiovascular)
  3. Diabetes $2173... 505 hits (diabetes)
  4. Trauma $2555... 4 hits (trauma)
  5. Back problems $1973... 4 hits (pain; not easy to search)
  6. Mental disorders $1857... 75 hits (behavioral|psychiatry|psychology|mental)
  7. Osteoarthritis $1648... 4 hits (arthritis)
  8. COPD/Asthma $1284... 15 hits (copd|asthma)
  9. Hyperlipidemia $871... 0 hits (hyperlipidemia|hypercholesterolemia|cholesterol)
  10. Hypertension $858... 4 hits (hypertension|"blood pressure")
Okay, so this is hardly a scientific study, right? In fact, it is a bit silly. But it backs my point that for chronic conditions where people seek out management and coping tools and supportive communities, mhealth buzz is circling these conditions. I can see why trauma might not have much here as it is more of an acute problem, albeit with chronic sequelae. I am surprised to see very few mentions of pain. HIV is not in the top 10 but has 502 hits.

Diving deeper into the mental health stuff, there are a few mentions of depression, anxiety, addiction and PTSD, but nothing on bipolar disorder (people often don't think of bipolar disorder even though it can be more costly than diabetes). There is an interesting talk on Tuesday evening at 5pm called "Open mHealth - Integrating mHealth Apps & Devices to Enable Better Health:"
"Imagine a world where a person with Post Traumatic Stress Disorder (PTSD) is able to share with their clinician--in real-time--their mood, behavior and medication data so they can collaboratively get to a richer, data-driven view of how they’re improving between clinical visits." 
I won't be able to make it to that one, unfortunately, but if you attend, please tweet or blog or comment here about it. (thanks!)

Sunday, October 14, 2012

Maryland Medicaid: Behavioral health problems predict skyrocketing medical hospitalizations

Earlier this summer the Maryland Department of Health and Mental Hygiene released four years of aggregated data on Maryland Medicaid patients and some measures of health care utilization, such as hospitalizations. The data can be found here (look at the July 11 meeting materials called Data Book). This is in the context of Maryland's attempts to integrate mental health, addictions, and somatic health (see Crazy Busy for more details).

The data show hospital admissions for the top 25 conditions -- such as stroke, diabetes, pneumonia, etc -- and further break up these data into four groups of people:

  • people who also have a mental health disorder
  • people who also have a substance use disorder
  • people who have both a mental health and a substance use disorder
  • people who have neither of these conditions (ie, just a somatic disorder, like asthma or diabetes)

Taking this data and analyzing it in more detail led to some amazing discoveries that are buried in the data tables. For example, in 2011, adult patients in Maryland Medicaid's HealthChoice program who had co-occurring mental health (MH) and substance use (SU) disorders were admitted to hospitals 800-1500% more often than those without these co-occurring conditions -- hospitalized 8-15 times more often for things like pneumonia, asthma, diabetes, epilepsy, and cellulitis.

That is worth saying one more time:
Patients with both mental health and substance use disorders were hospitalized 8-15 times more often for things like pneumonia, asthma, diabetes, epilepsy, and cellulitis.
Here is a graph demonstrating this amazing connection. People with co-occurring mental health problems were admitted 2-4 times more often for these 6 types of medical conditions, and people with co-occurring substance use problems were admitted 4-7 times more often. This is based on claims data submitted to Medicaid and analyzed by the Hilltop Institute. The data are normalized such those without mental health or addiction comorbidity have a relative risk of 1 for hospitalization

The six conditions are the primary diagnoses as indicated by the Medicaid data: epilepsy; HIV-related illness; cellulitis/septicemia; congestive heart failure; respiratory conditions (bronchitis, pneumonia, asthma, and COPD); and diabetes mellitis.

It's no wonder there is data that people with chronic mental health conditions treated in the public health system die 25 years earlier.

If there was ever a reason to better integrate mental health care, addiction treatment, and primary care, this is it.

Sunday, August 19, 2012

Crazy busy

So I noticed it's been 4 months since I last posted on HIT Shrink. Lots of things going on. I've had a few people ask me about what I've been working on, so thought I'd put it up here... if only for me to keep track of, including all the links to these areas.

  • Epic Implementation
    After about 15 months of planning, meeting, coding, optimizing, and preparing, we went live on Epic a week ago. We had Cerner before that, and a homegrown CIS before that. The initial CIS was mostly labs, rads, and dictations as I recall. With Cerner PowerChart and FirstNet, we added nursing documentation, MAR, and some clinical documentation, though we had a hybrid system that included a paper chart for all non-dictated physician documentation (mostly progress notes, as H&Ps, consults, and DC summaries were dictated) and paper physician orders. A handful of small, contained areas (Peds, OB, Psych, ED) did do CPOE, but expanding that to the rest of the hospital, and our many community-based physicians, was untenable.

    Now we have (mostly) abandoned paper documentation, with all physician documentation being done by either phone dictation, templates (using Notes and/or Notewriter plus/minus partial dictation), or Dragon. We are piloting Dragon Enterprise with a small number of physicians, otherwise they are using their own licenses. Still having network connectivity problems with wireless devices, especially BYODs -- particularly those of the Apple persuasion. And ironing out printing issues.

    Overall, the switch has gone amazingly well. I attribute this to excellent executive support, mandatory classroom (not online) physician training (8-12 hours), extensive clinician involvement in the build, and heavy at-the-elbow support.

  • Maryland's Behavioral Health Integration
    In 2011 we kicked off an effort to combine the Department of Health and Mental Hygiene's two administrations -- Mental Hygiene Administration and Alcohol and Drug Abuse Administration -- into an integrated administration which is expected to be called the Behavioral Health Administration. The Joint Chairman's Report lays out much of the rationale, as well as the challenges. Health Secretary Joshua Sharfstein MD MPH, previously Baltimore City commissioner, then a stint at FDA, has led an admirably open, transparent, stakeholder-rich effort to shape this integration effort.

    The past 3 months have been heavy with 2-to-3-hour long Workgroup meetings. I've been attending many of these meetings (maybe a third to a half), representing the Maryland Psychiatric Society. The main points I've been making mostly revolve around ensuring mental health parity and integration of primary care. Besides the large overall Workgroup, there have been four, more focused, workgroups (links have meeting agenda, minutes, resource documents). Their charges are as follows:
    • Systems Linkage:
      Purpose: To make a recommendation on those factors that should be present to promote "integration." For example, should there be a shared electronic health record among all providers within an MCO? What factors indicate “integrated” care, and what factors indicate “collaborative” care?
    • State/Local and Non-Medicaid WorkgroupPurpose: To make a recommendation on what services/financing should be left outside a “Medicaid” integrated care model to accommodate non-Medicaid eligible populations, or non-Medicaid-eligible services. This Workgroup will also make a recommendation on the roles that state and local government should perform depending on which services/financing are left outside of the Medicaid financing model, as well as how to support and interface with selected model.
    • Evaluation and Data WorkgroupPurpose: To determine what data is available and relevant to the ultimate recommendation on the model, and to make a recommendation on potential measures to evaluate any selected model.
    • Chronic Health Homes Workgroup
    • Purpose: To make a recommendation on a new “Health Home” service under the Affordable Care Act, and make a recommendation on how the new service could be developed to support any integration model. For example, this workgroup would help define the service; define the population eligible for the service; and define the provider qualifications to deliver the service.
  • Maryland Health Benefit Exchange
    Maryland is among the earliest states to respond to the ACA's requirement of health insurance exchanges that include qualified health plans (QHPs) which must include a minimum set of essential health benefits (EHBs). This process has been unfolding extremely quickly over the past three months, with several meetings per week across all the advisory committees, as well as the governing Board. I've served on the Plan Management Advisory Committee, which has completed its work and submitted a Summary to the Board and to the legislature. Over 200 pages of public comments have been submitted to the Board, which met last week. Ensuring that there are requirements to demonstrate Parity compliance, as well as adequate provider network standards, has been the main focus of mental health advocates.

    The other advisory committees include Navigator, Continuity of Care, Financing, and Exchange Implementation.
  • Maryland Health Information Exchange
    The MHCC Policy Board continues to slog through its policy discussions. The current list of policies is in a .pdf document here. We've also been working on draft regulations for HIEs in Maryland. Initial public comments on these draft regulations are available on the HIE website.

Other things I've been working on include APA's Committee on Electronic Health Records, M3Clinician (a clinician portal for the M3 mental health screening tool), and mental health parity. This last thing includes working with folks in the Parity Implementation Coalition, testifying at a Congressional parity hearing, and even appearing on the Diane Rehm Show.

I'll try to post a bit more often here, but you can also keep an eye on Shrink Rap, Clinical Psychiatry News, and a new column on health IT to start soon in Psychiatric News.
Thank you for reading.

Friday, April 13, 2012

Stage 2 MU: Universal Depression Screening?

In Clinical Psychiatry News yesterday, I wrote a column entitled, "Effect on Psychiatry of Stage 2 Meaningful Use," where I focused primarily on the proposal that all "meaningful EHR users" must collect and report 12 clinical quality measures or CQMs. This is not to be confused with the objectives that must also be achieved. Both the objectives and the measures have core items and menu items. The core ones are mandatory (unless one meets exclusion criteria) and the menu ones are selectives (eg, one must be chosen from a list of options).

In the column, I listed the proposed core CQMs (as listed in Table 6 of the Notice of Proposed Rule Making or NPRM) and focused particularly on one of the proposed CQMs, depression screening. Here is the proposed list for Stage 2:

  • receiving consultant reports
  • functional status assessments for patients with complex chronic conditions
  • controlled hypertension
  • medication reconciliation
  • use of high risk meds in the elderly
  • therapeutic drug monitoring
  • antithrombotic use in ischemic vascular disease
  • obesity screening and counseling in kids
  • tobacco use screening in adults
  • cholesterol screening in adults
  • depression screening for ages 12 and up

[For a refresher on Stage 1, this article should be helpful.]
CMS actually proposed two possible schemes for reporting these measures. One scheme -- called 1a -- is to select 12 measures from a larger list of 125 options in Table 8 of the NPRM, with at least one from each of 6 domains. The other -- called 1b -- is to report on all 11 of the Table 6 measures (listed above) and 1 from Table 8. The former scheme makes more sense for most specialists, especially psychiatrists, because it maintains flexibility and does not force us to report on things like cholesterol screening and antithrombotic use, both of which are most likely being addressed by a patient's PCP (primary care physician). If one instead reports CQMs via the Medicare Physician Quality Reporting System's EHR Reporting Option, then one can skip the above two schemes (at least, that's how I read it).

This Table 8 list of 125 clinical quality measures is the proposed list, not the final list that will likely be a subset of these based on public comments. Out of this long list, I list below those CQMs that seem to be most relevant to Psychiatry, while being sure to include at least one measure from each domain.

Initiation & Engagement of Alcohol & Drug Treatment
Clinical Process/ Effectiveness
Medication Reconciliation
Patient Safety
Major Depression: Diagnostic Evaluation
Clinical Process/ Effectiveness
Major Depression: Suicide Risk Assessment
Clinical Process/ Effectiveness
Anti-depressant Medication Management
Clinical Process/ Effectiveness
ADHD: Follow-Up Care for Children Prescribed ADHD Medication
Clinical Process/ Effectiveness
Mood disorders: Appraisal drug & alcohol use
Clinical Process/ Effectiveness
Bipolar Disorder: Monitoring change in level-of-functioning
Clinical Process/ Effectiveness
Screening for Clinical Depression
Population/ Public Health
Documentation of Current Medications in the Medical Record
Patient Safety
Depression Remission at Six Months
Clinical Process/ Effectiveness
Depression Remission at Twelve Months
Clinical Process/ Effectiveness
Depression Utilization of the PHQ-9 Tool
Clinical Process/ Effectiveness
Child & Adolescent Major Depression: Suicide Risk Assessment
Patient Safety
Dementia: Staging of Dementia
Clinical Process/ Effectiveness
Dementia: Cognitive Assessment
Clinical Process/ Effectiveness
Dementia: Functional Status Assessment
Patient and Family Engagement
Dementia: Counseling Regarding Safety Concerns
Patient and Family Engagement
Dementia: Caregiver Education and Support
Patient and Family Engagement
Dementia: Counseling Regarding Risks of Driving
Patient Safety
Closing the referral loop: receipt of specialist report
Care Coordination
Adverse Drug Event Prevention: Therapeutic drug monitoring
Patient Safety

So, if the 1b proposal is chosen by CMS, psychiatrists will have to report on cholesterol and anticoagulants, however this will also result in near-universal depression screening. This would likely result in a lot more referrals to mental health specialists when patients are found to score positively on the screening tool (note that any standardized screening tool is acceptable, such as the BDI, GDS, PHQ-9, and M3), resulting in improvements in morbidity and reductions in costs associated with undiagnosed and untreated depression.

But, it is also likely to result in more PCPs, pediatricians, and OB-GYNs managing mild and moderate depression. Unless they also screen for bipolar disorder, a certain percent of these people -- who have undiagnosed bipolar disorder -- will be placed on antidepressants and be harmed from induction of mania, hypomania, or a mixed state by the antidepressant. If this 1b proposal is chosen, we will need to ensure that PCPs know how to screen for bipolar disorder, and that they are able to recognize mania or hypomania induced by an antidepressant.

If the 1a proposal is chosen, then providers will be able to pick and choose which of the Table 8 CQMs (at least one from each domain) is chosen. While this will be preferable because specialists won't have to report on measures that they rarely address, this path will NOT result in widespread screening of depression, as I expect providers will pick those measures that they already record in most patients currently.

I must say that I am torn between the two options, as I would like to have my cake and eat it, too.

Please comment on your thoughts about this. Also, let CMS know your thoughts by providing public comment prior to May 7 at this page.

Friday, February 24, 2012

Meaningful Use Attestation by Psychiatrists

The ONC released a spreadsheet recently of all the eligible providers (Medicare and Medicaid) who attested to Stage 1 Meaningful Use (MU) of certified electronic health records (EHRs). I downloaded the results and did some pivot table magic to see which EHRs are being used by physicians identifying their specialty as "psychiatry."

There were 285 people who said they were compliant with Stage 1 MU. Interestingly, there were others who identified themselves in other ways. Three were "Community Mental Health Centers," even though HITECH specifically included them (but a doc there can attest, so I guess it's ok); they used Meditech and HCA. Five identified as "psychiatric" (Cerner, Design Clinicals, Siemens, and Wellsoft). And 16 as "psychiatric unit (Epic, Cerner, NextGen, McKesson, GE, HCA, Meditech, Midas+, Siemens). Here is data on the 285 identified as "psychiatry." If I get ambitious, I'll come back and hyperlink the names.
[edit 2/26: the links are in the Clinical Psychiatry News version of this article]

After discounting the large system products (eg, Epic, Cerner), you can get an idea about the EHR products being used in outpatient private practices.

Vendor Name EHR Product Name Total % of all Psychiatrists
Epic Systems Corporation EpicCare Ambulatory - Core EMR 129
EpicCare Inpatient - Core EMR 2
Epic Systems Corporation Total 131 46.0%
Practice Fusion Practice Fusion 22
Practice Fusion Total 22 7.7%
Cerner Corporation Health Sentry 8
Millennium Powerchart, Healthe Exchange, IQHealth, Health Sentry, Cerner Health 1
Powerchart and Cerner Healthe 8
Cerner Corporation Total 17 6.0%
Allscripts Allscripts ED 1
Allscripts Enterprise E HR Modular 2
Allscripts Enterprise EHR 2
Allscripts ePrescribe 1
Allscripts MyWay EHR 1
Allscripts PeakPractice 1
Allscripts Professional EHR 5
Allscripts Total 13 4.6%
Community Computer Service, Inc. MEDENT 9
Community Computer Service, Inc. Total 9 3.2%
Valant Medical Solutions, Inc. Valant Premium Psychiatric Suite 9
Valant Medical Solutions, Inc. Total 9 3.2%
DrFirst Rcopia MU 8
DrFirst Total 8 2.8%
MedSeek, Inc. eHealth ecoSystem 8
MedSeek, Inc. Total 8 2.8%
WellCentive WellCentive Registry 8
WellCentive Total 8 2.8%
eClinicalWorks LLC eClinicalWorks 6
eClinicalWorks LLC Total 6 2.1%
ICANotes, LLC ICANotes EHR/EMR for Behavioral Health 6
ICANotes, LLC Total 6 2.1%
NextGen Healthcare NextGen Ambulatory EHR 5
NextGen Inpatient Clinicals 1
NextGen Healthcare Total 6 2.1%
McKesson Medisoft Clinical 4
McKesson Total 4 1.4%
e-MDs, Inc. e-MDs Solution Series 3
e-MDs, Inc. Total 3 1.1%
MedPlus, A Quest Diagnostics Company Care360 EHR 3
MedPlus, A Quest Diagnostics Company Total 3 1.1%
UNI/CARE Systems, Inc Pro-Filer(TM) 3
UNI/CARE Systems, Inc Total 3 1.1%, Inc. Amazing Charts 2, Inc. Total 2 0.7%
Greenway Medical Technologies, Inc. PrimeSuite 2
Greenway Medical Technologies, Inc. Total 2 0.7%
MedCPU Inc. Meaningful Use Advisor 2
MedCPU Inc. Total 2 0.7%
Meditab Software, Inc. IMS 2
Meditab Software, Inc. Total 2 0.7%
MTBC (Medical Transcription Billing Corporation) MTBC-EHR 2
MTBC (Medical Transcription Billing Corporation) Total 2 0.7%
Nuesoft Technologies, Inc. NueMD EHR 2
Nuesoft Technologies, Inc. Total 2 0.7%
Office Ally EHR 24/7 2
Office Ally Total 2 0.7%
TheraManager LLC MaestroMed 2
TheraManager LLC Total 2 0.7%
Waiting Room Solutions 2011 Waiting Room Solutions Web Based EHR and Practice Management System 2
Waiting Room Solutions Total 2 0.7%
ADP AdvancedMD AdvancedMD EHR 1
ADP AdvancedMD Total 1 0.4%
athenahealth, Inc athenaClinicals 1
athenahealth, Inc Total 1 0.4% Inc. drchrono EHR 1 Inc. Total 1 0.4%
GE Healthcare Centricity EMR 1
GE Healthcare Total 1 0.4%
HealthFusion MediTouch EHR 1
HealthFusion Total 1 0.4%
Henry Schein Medical Systems MicroMD EMR 1
Henry Schein Medical Systems Total 1 0.4%
Ingenix Ingenix CareTracker 1
Ingenix Total 1 0.4%
IOS Health Systems Medios 1
IOS Health Systems Total 1 0.4%
iSALUS Healthcare OfficeEMR 1
iSALUS Healthcare Total 1 0.4%
Medical Informatics Engineering WebChart EHR 1
Medical Informatics Engineering Total 1 0.4%
SuiteMed Intelligent Medical Software (IMS) 1
SuiteMed Total 1 0.4%
Grand Total 285

Hmm, the numbers are cut off. I placed this here in a Google Doc so you can see all the data.