I am back from attending the National Association of State Mental Health Program Directors NASMHPD Research Institute's18th annual conference of State Mental Health Agency Services Research, Program Evaluation and Policy in Arlington VA. I posted notes on all of the presentations that I attended so you can read about those as you like. They also promised to have slides available on their web site so you can get more detail as well (the abstracts are already available from the link above).

NRI was interested in getting more participation from research institutes like MIMH and offered us an all expenses paid trip to the meeting (as well as individuals from two other Institutes). As the newest Division Director at the Institute I got asked to attend. While NRI paid my way they did not ask me to write about the experience and they may have never even considered that someone would blog their conference. So the comments (and errors in those comments) in this and the other posts from the meeting are my own.

It was interesting that while NRI paid for our attendance, they made no efforts during the meetings to meet me (or I presume the other research institute attendees) or to have us meet together to explore their goals for our participation. While I was exhausted from the rounds of presentations this seems like a missed opportunity that I suggest they address if they do this again.

When I first looked at the two and a half day program there was lots of stuff that seemed of interest. After all, I have been working in public mental health for over 25 years so how could presentations for the NRI be uninteresting. The night before I read the abstracts and plotted out my selections. However, I was beginning to get cool feet and felt that I might be in for the usual sort of conference "fluff" that was long on flash and short on substance. One of the reasons for this is that a couple of the presentations were reporting anew the same sort of work I had read 25 years ago.

Well I was wrong. The presentations were anything but fluff. They were solid, data supported, presentations that showed states effectively using information to make changes to improve the lives of their citizens. Outstanding!

I have always been an "L" and believe that government could (and should) be a source of help for those in need. As a single individual I cannot have much impact of the mental health needs of my fellows. But as a member of a community, which is what a state represents, our united efforts and dollars can have a positive impact. Those attitudes have taken a beating in recent years, but the programs presented at this conference and the states hosting them have shown this can be true. They have helped excite me and restore my belief in the power of the community, as represented  by our governments, to be a positive force in the lives of those in need.

To me, this is great outcome and well worth my time and effort. Thanks for the invitation, I'll be back!

To make it easier to find them, here are the available posts:

• NRI - Better but not well
• NRI - Learning Collaboratives in New York
• NRI - Chronic Medial Disorders with SMI
• NRI - Behavioral health assessment system
• NRI - Collaborative Research
• NRI - Improving data
• NRI - PTSD
• NRI - NOMs and TRAC data systems
• NRI - Enhancing mental health treatment effectiveness
• NRI - Grading the states
• NRI - Describing state mental health agencies
• NRI - Evidence based practice in practice

From NRI's State Mental Health Agency Services Research, Program Evaluation and Policy in DC, PTSD Then and Now, There and Here, Robert J. Ursano, MD, Director, Center for the Study of Traumatic Stress www.usuhs.mil/csts.

Robert Glover (Ex Dir., NASMHPD) introduction. NASMHPD has been focusing on this at a high level and considers it a serious problem. Concerned about suicide prevention particularly with returning National Guard and reservists. 1-800-273-talk (www.shicidepreventionlifeline.org) suicide prevention lifeline which has access to military. On last Saturday 238 vets called the line for help!

"Trauma is bad for your health" and "War is a trauma."

Hospitalizations in the DoD for trauma are second only to those for pregnancy.

Trauma associated with death is important for some individuals but trauma associated with separation and return is the common event for all service personnel.

War is an intentional and human made traumatic event the combination of which create the most trauma. 1.5 million Americans have returned from Iraq or Afghanistan (300,000/state; 1,500-6,000 cases of PTSD/Depression per state). Because combat is an ongoing traumatic event it is harder to treat and does not respond as readily to medications.

Of the vets who have problems only 20% actually get specialized MH treatment. If you have PTSD you are likely to fear admitting the problem or getting service because it would negatively impact your career. This is no different from the risks people in the general population feel.

PTSD treatment (for all individuals) often does not begin for 12 years after the events. Even then only 65% actually get treatment. PTSD seems to increase over the course of time with injured solders having 12% PTSD and 9% depression after 7 months.

From NRI's State Mental Health Agency Services Research, Program Evaluation and Policy in DC two presentations on the implementation of evidence based practices.

Linking Outcomes to Services: Using Montana's Recovery Markers for Program Development and System Change, Alison Hwong & Bobbi Renner.

Fourth largest state but a rural frontier population with 12 Native American tribes and seven reservations. They have developed a web based system to measure their outcomes.

Four recovery markers are used

1. Employment status,
2. housing,
3. level of symptom interference, and
4. substance use.

which are collected every 3 months. The Information system also links to pharmacy data, demographic, diagnosis, functioning, living status, and service provides. Will be able to link to the medicare/medicaid data.

[The nice thing about the markers is that they are typically single items with a few categories. This will be simple for users to enter and conceptualize, though psychometrically weak. Their simplicity may also lead to system gaming.]

Problems with implementation

• Lack of computer experienced (one user called reporting a strange moving line on their screen. After a bit of description it turned out to be the cursor.)
• Communication problems between users and computer professionals
• Art vs. science [practitioners think about treatment as a holistic interpersonal enterprise while computer folks are interested in the change in recovery markers]
• providers say they were required to enter to much information without additional compensation (advocacy groups wanted more information)
• maintenance level
• providers felt that the information collected "will never be useful to us"

[The implementation issues discussed in this presentation are exactly the same as those addressed during the early days of computer applications in mental health. Why is it we keep rediscovering the same facts over and over? Why have the folks who develop these systems not been taught the lessons we have learned?]

Ohio's Wellness Management and Recovery Program: Working Collaboratively to Develop, Implement and Evaluate a Statewide Recovery Initiative, Marion Becker, Airia Sasser, Timothy Boaz, Amber Gum, Wesley Bullock, Kelly Wesp, Deborah Wilcox, & Stephanie Rich. www.wmrohio,org 

Participants develop a wellness plan using a psycho educational (social learning model) curriculum with role plays, video, lots of interactive content. Trainers are pairs one consumer and one staff. Doing a rigorous evaluation because they want it to be an evidence based practice.

Session topics (10 2-hour sessions)

1. Mental Health recovery
2. Wellness
3. Understanding of Mental health
4. role of Medication in recovery and wellness
5. Learning to manage symptoms and side effects
6. Effective communication
7. communicating with your providers
8. Coordinating your care
9. Building social supports and involving others
10. Planning for wellness.

Lots of focus on learning new skills.

Have moved from systems data to individualized data. Asking each consumer what they want to achieve. Steve will provide these on request.

[This is a really nice idea and one that I have been advocating for a number of years. If an individual's services are being directed to areas A, D, and E why would we bother assessing changes in areas B, and F? While this makes sense to evaluators this does not seem to be clinically reasonable. With our advanced statistical and data management techniques we should be able to define specific outcomes for specific individuals rather than assuming everyone is the same.]

From NRI's State Mental Health Agency Services Research, Program Evaluation and Policy in DC, Understanding State Mental Health Agencies, Ted Lutterman & Bernadette Phelan  (NRI).

State mental health agencies

Agencies provide services, public health, and public safety.

There has been a shifts in organization with reductions in the number of states with a separate department (19-13 1981-2007) and fewer linked directly to the governor. This is important because the closer to the governor the more resources are available.

Most states (26) have MH and ADA in the same agency. This has increased since the 80s. Most directly fund agencies (28) but many fund counties/cities (16) which adds another level of government to the provision of services.

In 1954 there wee 354 state hospitals with 550,000 residents. In 2007, 228 hospitals with 50,000 residents. The size of hospitals dropped in the 60s and 70s but hospitals themselves did not start closing in large numbers until the 90s.

In 1979 1 our of 3 hospital beds were psychiatric. In 2007 1 out of 7 is psychiatric. These decreases have lead to shortages of acute psychiatric inpatient capacity with increase ED treatment, long treatment waits, overcrowding, etc. States have been addressing this by trying to increase programs to keep folks in the community.

Many states are working to decrease service fragmentation by working with various other state agencies (housing, employment, juvenile and criminal justice, etc.

Most states have mental health courts or diversion programs (43).

Public mental health agencies treat 2% of the population with 96% of those being treated in the community. Almost half of these are not in the workforce and most live in private residence (76%). Most of those (75%) treated are seriously mentally ill.

Over time there has been no improvement in terms other the number who return to the hospital within 30 days (9%) or after 189 days (20%). [Interesting that with the increases in treatment effectiveness that this has not changes. Though it may be related to the fact the the number of individual served has decreased so that those in the hospital will likely have more difficulties.]

There has been a shift in funding from state funds to medicaid. with 45% currently coming from the latter.

Other state agencies

MH services are also provided by other state agencies. NRI is looking to see if they can get information from these other agencies so that they can better understand what states are doing for mental health. Sending letters from the state MH agencies and identifying where they can find the answers to the questions.

NRI has no data to report because they need to get the appropriate clearances (will report at the next year's conference).

While the states and agencies are interested in this there are challenges in getting the data. Agencies may not record data in an obvious way. For example states working with education counted all of the $s spent on special education as mental health services.

Note that all of the data they report is publicly available.

From NRI's State Mental Health Agency Services Research, Program Evaluation and Policy in DC, Grading the States: Assessing State Mental Health Systems, Michael Fitzpatrick & Laudan Aron (NAMI) www.nami.ogg/grades

NAMHI ad ranked states previously but decided grades would be fairer to the states. They also offered suggestions of ways to move from a lower to a higher grade in each area. The report is aligned it with the President's New Freedom Commission recommendations and the transformational language it embodies. Will be doing this biannually with hopes of doing it annually.

In the 2006 report the national average D

Positive trends

• Increase focus on evidence based services
• more consumer/family involvement
• increased linkages to other departments and partners
• 'pockets of innovation in virtually every state, even those receiving an "F"'

Negative trends

• Budget cuts or flat funding
• Medicaid changes (e.g., eligibility, cost-sharing)
• Shortage of inpatient beds
• Cost shifting to other systems
• Long waiting lists for community services

They are now working on the 2008 survey trying to refine the methodology and gather information from additional systematic databases. The presenters took a lot of feedback and welcomed more. So if you have suggestions about how to improve their methodology shoot them an email.

From NRI's State Mental Health Agency Services Research, Program Evaluation and Policy in DC, Enhancing the Clinical Effectiveness of Mental Health Treatments: Research to Improve Practice & Inform Policy, Philip Wang, (NIMH).

What are the unmet needs

• Mental health treatment seeking is very slow 10-30 years
• About 41% needing care receive it.
• The sorts of treatments being received are sub optimal
• For people with SMI only 20% re receiving minimally acceptable care

How to improve effectiveness (innovation and enhancement)

• Because the cost of genetic typing costs have dropped so much we can use that information to identify diseases with specific genetic locations
• Increase potential for more personalized treatments
• If the Quality adjusted year costs are$50k or less it would be appropriate to do these procedures. This is because dialysis costs this much and the government is willing to pay for that treatment.
• There is an increasing use in new medications (even before comparative studies) which has resulted in rapidly increasing costs.
• States are spending a large percentage of their mediation budgets on atypical antipsychotics event thought do not know if this is appropriate or not.
• While there is a lot of money being spent on healthcare there is no evidence that this improves health outcomes. [Suggested that this means we need more effective treatments. What it really means is that we are leaking money in our systems which are not as efficient other nations.]
• Use real use data
• use quasi-experimental studies
• Use simulations to create studies

From NRI's State Mental Health Agency Services Research, Program Evaluation and Policy in DC two presentations about new data systems NOMs and TRAC.

Are We Ready for the NOMS?, Mark Reynolds and Tracy Leeper (OK). http://www.odmhsas.org/nri

Spent a lot of effort collecting data and using it but never sent it back to the providers. National Outcomes Measures. Web based fee for service model which requires good admission data to stat billing.

Their admission-discharge data did not show much improvement. This is because the discharge data is never updated providers simply carry data over from the admission.

Changes to address this problem

• 6 month updates
• after 6 months of no service unplanned discharge (this will attract the agency's attention as they get dinged for this difficulty).
• clinician of record for each client and their performance will be tracked
• required very few data elements (6) for an update

They provide a very nice set of reports for the agencies. Go to agencies and talk with clinicians about the data that they are providing. They also train their own staff how to understand the data and reports. 

TRansformation ACcountability (TRAC): A Coordinated Approach to Performance Measurement by the Center for Mental Health Services, Diane Abbate and Jessica Taylor. https://www.samhsa-gpra.samhsa.gov/home/index.htm

Performance management and outcomes systems for all CMHS programs (PRNS, CMHI, PATH, PAIMI). Needed to meet SAMHSA data reporting requirements with standardized measures and accountable. TRAC is basically reporting the NOMs (service) data. Data is collected at baseline, reassessment (6 or 3 months), and discharge. They are collecting client level data though it is deidentified. There is no mechanism for uploading data require the individual data to be entered one at a time.

From NRI's State Mental Health Agency Services Research, Program Evaluation and Policy conference, Collaborative Research Opportunities for Addressing Mental Health Policy Issues: Building partnerships between state mental health authorities, NIMH, and SAMHSA, Renata Henry, Michael Schoenbaum, Jeanne Rivard, David Chambers, and Jeffrey Buck

R. Henry (State perspective) - States have to make decisions but there are gaps between research, policy, and practice so that decisions are made without adequate information. The collaboration helps to identify key questions, disseminate relevant research, and interpret and use data to inform decisions. This important because information is shared across states providing for faster uptake and improved management of information.

D. Chambers (NIMH) - While there has been an increase in research on EBP policy in the states there was a lack of coordination of information to decide what impact these changes have made. Want to explore using existing state and national data to understand the impact of policy. Transformation grants are trying to push integration of the various data streams to create added value. Want to fund collaborative grants with 2 PIs State and Researcher). Want to understand the total impact of policy changes across a variety of different agencies within the state (e.g., MH, corrections, SSI)

J. Rivard (NRI) - State Mental Health Policy Laboratory. This is really a data set that can be queried to look at different policy changes across states. The goal is to gather data that is specific to policy not to look for policies within data. Potentially support:

• Multi-state interest groups based on current efforts and policy
• States with data warehouses can help move this forward with cross agency data
• Identify states with specific initiatives and look for common data elements for cross state analyses.
• Identify common data elements across all states in national data sets for policy analysis.

There is currently a limit in that we collect data from the mental health system but NOT from all of the mental health treatment that occurs in other agencies and sectors. Nor is there very good sharing of data and research between any of the stakeholders.

J. Buck (SAMHSA) - Sees collaboration and a broader collection and sharing of data is something that is going to be very important. In the next couple of months will be releasing their data strategy (be more systematic about data collection and analysis, need to look at issues across various agencies). Want to increase data interoperability in administrative data sets being able to join data at the level of the consumer and provider.

From NRI's conference on State Mental Health Agency Services Research, Program Evaluation and Policy in DC, Improving Data Interoperability: Opportunities for States, Jeffrey Buck, & Rick Friedman.

J. Buck (SAMHSA/CMHS) - Data interoperability is the degree to which databases can be combined to provide information on aggregate data questions. These are affective by coding standards, payment method, and reporting elements. The difficulty comes from the siloed systems and from the constant revisions of those systems. (Hard to find the same person or provider in multiple systems.) Interoperability is a goal in the larger policy issues (e.g., New Freedom Commission on Mental Health and the IOM quality report). The focus is to encourage the development and use of client-centric data systems using a common data platform. Confidentiality (HIPAA and 42 CFR Part 2) is important but these do permit data sharing under certain circumstances (e.g., research, audits, healthcare operations) and SAMHSA is developing a guide to confidentiality.

Current status. Good-Most states use unique client identifies with client-level data, encounter/claims data and the majority link with one or more agencies. Bad-Multiple IT platforms, much legacy systems, many use state-owned or proprietary software, unique service coding and DSM IV, few have have detailed prescription drug data.

How to evaluate interoperability. some criteria include:

• Client ID
• Provider ID
• Access common and unique data elements
• Assess EHR "readiness" (how much can information can be easily shared, currently working on a set of measures for this)

To move toward interoperability you need to invest effort and FTE to move toward this goal. Consideration needs to be given to the process because this is an expensive and difficult task. But early results are important to help develop and maintain momentum. Don' wait until you can get it all start small and expand. Start using of-the-shelf data systems. Implement HIPAA compliant coding and pay attention to the provider identifiers.

[It is interesting that there was a strong call for an open architecture. In Missouri there was an attempt to follow this approach but it has devolved into an expensive and over due customized system.]

R. Friedman (Center for Medicaid and Medicare, CMS) - Medicaid Information Technology Architecture (MITA) and Mental Health. MITA will give us some information of where SAMHSA is headed. Focus more on outcomes and as a collaborative process. Promoting sharing of data among partners, access to multiple databases. Does not dictate structural changes in individual states. MITA is a web based modular development system that states can use to develop their own system but based on shared standards. Provides the basis for EHR, eRx, PHR, ...

Components:

• Business architecture (operations concept, maturity model, business process model, self-assessment, business services)
• Information/Data architecture (
• Technical architectures

Asking states to complete a State Self Assessment (SS-A) to identify where the state is in terms of their maturity (interoperability) and where they want to go. This helps the feds identify where a state is so that they can share that information telling the state to whom to go talk with who already knows how to do where you are trying to go. This is not yet available for behavioral health but will soon be doing pilot work on this assessment in the real world. There was a point raised during the discussion that compared the SS-A to a readiness to change model where there is a clear road map from where you start to where you could end.

CMS is willing to invest in systems that support enterprise-wide IT initiatives and pays 50-90% based on how much the medicaid  providers drive the process.

In my continued report from NRI's State Mental Health Agency Services Research, Program Evaluation and Policy in DC I am listening to the plenary presentation, Mental Health Policy 2008: Winners Curse?, by Richard G. Frank, PhD, Professor of Health Economics in the Department of Health Care Policy at Harvard University. The presentation was mostly taken from his book with Sherry A. Glied entitled Better But Not Well: Mental Health Policy in the United States Since 1950.

Starting with "scripture" from JF Kennedy from a 1996 presentation on mental illness and early diagnosis in the community, minimizing reliance on institutions, "restore and revitalize" their lives. The issues of the 60s are than same as today

1. Mental Health care and the well being of individual has improved notably
2. There is a long way to go
3. The forces that helped this happen are now presenting create new challenges and threats
4. Institutions aimed at creating a new stewardship of mental health represent a new frontier
5. What research needs to be done to underpin these new relationships

Things have gotten better

• Access to care - across various levels of mental health problems there has been a 62-66% increase in the rates of treatment.
• Quality of care - Individual getting appropriate medication related treatments have increased from to 20% to 50% range.
• Independent living - substantially more people with severe and persistent mental disorders are living in the community (about 75-92%). However there has been a large increase in the number of these people in correctional facilities and homeless.

There has been a shift in financing

• Change in costs - Have peen spending about 1% on public health care while we have moved private care 12% to 20%. So we have fallen behind because the costs have increased but the coverage has not.
• Changes in policy have not been driven by mental health policy but by social programs that are not designed for mental health (Medicaid, TANF, SSI, SSDI, child welfare) but they have been driving the changes. If you want to do a good job supporting mental health you need to understand all of these other programs. This means that there is often a lack of good mental health expertise, voice, and accountability in these various non mental health related programs.

How do we create new policy-leadership

• Consolidate under a single agency that deals across agencies. New Mexico is ding this. Very difficult!
• Create a matrix organization consulting versus action function (Mass)
• Bring MH into program agencies. Tends to quickly marginalize the MH folks.

Research

• What are the incentives and how might we change them to support the alignment.
• What organizational, financial and programmatic features are likely to work in different state political models. Which will be different for each state. 
• Need qualitative research to understand "what the rules of the game really mean."
• Investment need to me made in data systems so that we can understand what is happening. This suggests that the information MUST be shared across a variety of agencies.

It seems to me that the underlying message is that the changes in financial and policy have left the mental health field behind. We need to address this problem by developing systems and expertise that allow us to work within these new policies and to make mental health a "player" to share our experience to improve these policies.