Feed on
Posts
Comments

(Caution: This post is a draft of what I hope will be an article or book chapter – perhaps in the book on mental health planning I am trying to write.  Please be warned that it is a work in progress that may be changed to correct errors or for stylistic reasons.  Also, please do not use material from this post without proper attribution to the post or the references cited.)

CYCLES OF REFORM AND FAILURE AND THE NEED FOR BETTER PLANNING 

“They were morning glories — looked lovely in the mornin’ and withered up in a short time, while the regular machines went on flourishin’ forever, like fine old oaks. Say, that’s the first poetry I ever worked off. Ain’t it great?”

George Washington Plunkitt, New York Tammany Hall Leader, on Reforms, (Gunnison 1966)

Public Domain Photo

The history of the public mental health system for adults with serious mental illness has been marked by cycles of planned reform followed by failures to reach desired goals and subsequent public disillusionment. (Bachrach 1979; Goldman and Morrissey 1985; Sharfstein 2000; Grob and Goldman 2006; Pirkis, Harris et al. 2007) . Each cycle is a response to policies and interventions of the previous cycle, emerging technologies and contemporaneous societal trends. (Note: This analysis focuses on the adult mental health system.  Although a similar analysis might be useful for the child mental health system, it is beyond the scope of this paper.)

In 1985 Goldman and Morrissey counted four mental health cycles of reform (Goldman and Morrissey 1985):  The first in the early 19th century introduced moral treatment and focused on the asylum; the second in the early 20thcentury was associated with the mental hygiene movement and emphasized the psychopathic hospital; the third in the mid 20th century developed out of the community mental health movement and supported community mental health centers, and the fourth, in the 1980s came to be known as the Community Support Program and focused on a broad network of community-based mental health and social welfare services, which were coordinated with or absorbed into the mental health services (Goldman and Morrissey1985).  I would argue that after the Community Support Program we experienced a fifth cycle of reform – the consumer movement focusing on consumer involvement in all aspects of mental health care, the era of the new federalism, focusing on state mental health systems, and a sixthcycle, the era of evidence-based practices, the focus of which is the individual service.

Table 1, based on work by Goldman and Morrissey (1985), Palmer (Palmer No Date) and the Minnesota Psychiatric Society (2004) shows a more detailed and extended history of major national reform efforts from 1840 to the present and our formulation of the cycles of reform with which they are associated. We have re-named Goldman and Morrissey’s (1985) third cycle of reform beginning in 1955 “Deinstitutionalization and Community Mental Health.” Deinstitutionalization preceded community mental health, but the two eventually became linked in planning and service system development (Koyanagi 2007). We also have added three additional cycles of reform, the consumer movement, focusing on consumer directed care and consumer operated services (Fisher, Chamberlin et al. 2005), the New Federalism, focusing on state mental health systems (Osborne 1990) and the most recent, Evidence-based Practices emphasizing discrete services (Drake, Goldman et al. 2001).

Table 1 Brief History of National Mental Health Reform Efforts
First Cycle of Reform: Moral Treatment
1840: There were only eight “asylums for the insane” in the United States. Dorothea Dix
crusaded for the establishment or enlargement of 32 mentalhospitals, and transfer of those with
mental illness from almshouses and jails. First attempt to measure the extent of mental illness
and mental retardation in the United States occurred with the U.S. Census of 1840, which
included the category “insane and idiotic.”
Second Cycle of Reform: Mental Hygiene
1900: The “mental hygiene” movement began; Clifford Beers, a mental health consumer, who
shocked readers with a graphic account of hospital conditions in his famous book, The Mind that
Found Itself
1946:: On July 3, President Truman signed the National Mental Health Act, creating for the first time in US history a significant amount of funding for psychiatric education and research and leading to the creation in 1949 of the National Institute of Mental Health (NIMH).
Third Cycle of Reform: Deinstitutionalization and Community Mental Health
1955:: Congress authorized the Mental Health Study Act of 1955 and called for “an objective,
thorough, nationwide analysis and reevaluation of the human and economic problems of mental health.” The act results in the historic study conducted by the Joint Commission on Mental Illness and Health, Action for Mental Health.
1961: Action for Mental Health transmitted to Congress. It assessed mental health conditions and resources throughout the United States “to arrive at a national program that would approach adequacy in meeting the individual needs of the mentally ill people of America.”
1963: President Kennedy proposed and signed Community Mental Health Act that started community mental health center movement to substitute comprehensive community care for custodial institutional care.
1965: The CMHC (Community Mental Health Center) Act Amendments of 1965, (P.L. 91-211), were enacted and included the following major provisions: Construction and staffing grants to centers were extended and facilities that served those with alcohol and substance abuse disorders were made eligible to receive these grants. Grants were provided to support the initiation and development of mental health services in poverty-stricken areas. A new program of grants was established to support further development of children’s services.
1975: The CMCH Act Amendments of 1975 (P.L. 94-63) mandated a more detailed community mental health center definition emphasizing comprehensiveness and accessibility to all persons regardless of ability to pay, through the creation of a community governing board and quality assurance. Required core services expanded from the 1963 levels from 5 to 12, which included the following: Children Services Elderly Services Screening Services Follow-up Care Transitional Services Alcohol abuse Services Drug abuse Services.
1978: Medical Assistance (MA) added for community MH services (outpatient and day treatment).
Fourth Cycle of Reform: Community Support Program
1980: The Mental HealthSystems Act, (P.L. 96-398), restructured the federal community mental healthcenter program by strengthening the linkages between the federal, state, and local governments. The Act was the final result of a series of recommendations made by President Jimmy Carter’s Mental HealthCommission. Per the Mental Health Systems Act, a number of grant programs mandated for the CMHCs to assist in expanding services to meet an array of priority populations. They included the following:
* An expansion grant for a wide range of services for the severely mentally ill (SMI) population;
* Grants for the severely emotionally disturbed (SED) population;
* Non-revenue producing services were also funded via a grant aimed at expanding education and consulting needs;
* Additionally, the commission sought to include consumer input and involvement in service and treatment.
Fifth Cycle of Reform: Consumer Movement
1970: Founding Mental Health Insane Liberation Front in Portland OR.
1978: Judi Chamberlin publishes On Our Own :patient-controlled alternatives to the mental health system.
1995: First Substance Abuse and Mental Health Services Administration Center for Mental Health Services Consumer Affairs Specialist hired.
1996: Passage of the first parity law. The law prohibited insurers or plans serving 50
or more employees from setting lower annual or lifetime dollar caps on mental health benefits
than for other health benefits.
1999: : The Supreme Court issues its opinion on Olmsteadv. L.C which held that it is a violation of the Americans with Disabilities Act to keep individuals in restrictive inpatient settings when more appropriate community services are available.
Sixth Cycle of Reform: New Federalism
1981-2: Alcohol, Drug Abuse and Mental Health Block Grant, replaces Federal Mental Health Systems Act which is repealed. ADMHS block grant decreased by 30% resulting in dramatic service reductions. Despite passage of block grants, the federal share of funding decreased to 11% of the total while state and local funding share increased.
1986: Mental Health Planning Act of 1986 (Federal law requiring state plans) passed; Case management established as a distinct benefit under Medicaid; Medicaid amendments improve MH coverage of community MH services, add rehabilitative services, and expand clinical services to homeless.
1988: Behavioral health managed care introduced. Massachusetts was the first state that utilized a managed care platform regarding service of its behavioral healthcare needs.
The Clinton White House held a conference on mental health issues in June 1999 that focused on dispelling the myths about mental illness and decrying prejudices against behavioral health consumers, one of which was insurance coverage that excludes behavioral health services. The conference also brought together the mental health community in anticipation of the Surgeon General’s Report. Mental Health: A Report of the Surgeon General published in late 1999 seeks to eradicate the stigma surrounding mental health and simultaneously encourage the use of innovative pharmaceutical and psychotherapy treatments
2002: President Bush forms the New Freedom Commission on Mental Health, which will seek “to conduct a comprehensive study of the United States mental health service delivery system, including both private and public sector providers.” The Commission is charged with a set of objectives that includes reviewing the current quality and effectiveness of private and public providers, identifying innovative services, treatments, technologies, and issuing a report on its subsequent recommendations.
2003: New Freedom Commission final report issued.
Seventh Cycle of Reform: Evidence-based Practices
2007: National Registry of Evidence-based Programs and Practices (NREPP) for mental healthservices, substance abuse treatment and substance abuse prevention launched by Substance Abuse and Mental Health Services Administration. Efforts generally influenced by work of Cochrane(Cochrane 1972); in public mental health by Drake and colleagues (Drake, Goldman et al. 2001). 
End of  Table

None of these reforms have lived up to expectations (Bloche and Cournos 1990; Bell and Shern 2002), although some recent analyses assert that incremental gains can ensue from these reforms (Koyanagi and Goldman 1991). As Goldman and Morrissey wrote of the first three:

“Each [of the first three reforms]… failed to eliminate chronicity or to fundamentally alter the care of the severely mentally ill. In each cycle, the optimism of reform gave way to pessimism and therapeutic nihilism toward the increasing numbers of incurable chronic mental patients. In the face of an expanding population of needy patients, public support turned to neglect (Goldman and Morrissey 1985, p. 726).

The Community Support Program and the New Federalism also have generated disillusionment (Bloche and Cournos 1990). This disillusionment stems from phenomena such as “transinstitutionalization,” or the “migration” of persons with mental illness to nursing homes, jails, prisons, or homeless shelters, fragmentation of services, decreasing real dollars in entitlement streams and “formidable obstacles to eligibility’ (Bloche and Cournos 1990): These problems resulted in commissions, reports and program initiatives (Bell and Shern 2002; New Freedom Commission 2003) we have discussed in previous posts and will discuss in future ones.

It can be argued that cycles of reform five and seven are still in process, but they are already encountering problems. The consumer movement continues to seek a unified voice and evidence for its positions ((Fisher, Chamberlin et al. 2005)); and states, counties, and localities are having difficulties implementing evidence-based practices with fidelity and taking them to scale (Frank, McGuire et al. 1999; Drake and Bond 2008). As Drake and Bond (2008)note:

“…it is clear that most states lack the infrastructure for [implementing EBPs] (NAMI 2006). Instead, salaries and training continue to decrease, states spend their limited training dollars on conferences and brief trainings that do not change clinicians’ behavior (Hoge et al. 2003), and services continue to erode (New Freedom Commission on Mental Health2003). Implementing effective services always takes a back seat to financial issues, as agency administrators cautiously weigh financial risks and benefits (Panzano and Roth2006). Beyond implementation, states do not have the capacity for on-going fidelity monitoring, providing technicalassistance, and aligning financing with evidence-based practices—all of which would be needed to sustain and amplify success (p.225).

 This history of repeated and disappointing reform raises the question of why we do not do better. We can point to many factors that limit the effectiveness of our reforms like resource constraints and the lack of “magic bullet” technologies to cure mental illnesses or fix important problems in access and quality. However, my analysis, explained in detail in forthcoming posts, is that a primary reason why mental health system reforms have not been more successful and satisfying is that they have been inadequately planned leading to less than optimal resource deployment and unrealistic expectations (Pirkis, Harris et al. 2007).

Given this analysis, my recent work has been focusing on how to improve mental health system planning. The federal government, states, counties, localities and providers repeatedly make mental health systems plans. Some of these plans, such as the National Plan for the Chronically Mentally Ill (United States. Dept. of Health and Human Services. Steering Committee on the Chronically Mentally Ill. 1980) are developed by Presidential and other commissions. Other plans such as Olmstead Plans (National Council on Disability 2003) are required by the judiciary.  Additional plans, such as Block Grant Plans (SAMHSA’s National Mental Health Information Center: Center for Mental Health Services 2010) and Comprehensive Mental Health Plans (Substance Abuse and Mental Health Services Administration, 2010) are necessary to receive funding. Still others, such as 10 year plans to end homelessness (National Alliance to End Homelessness 2000) are recommended to guide policy and service development. Also,  plans such as state budget plans and transformation or reform plans (e.g., North Carolina Department of Health and Human Services 2007) are intended to satisfy legislative requirements and so on.

Better planning for mental health systems will always be in the critical path of deploying new technologies and resources, can result in safer and more effective mental health systems and can prepare stakeholders for the risks and limitations of reform as well as its opportunities. Nevertheless, to our knowledge there has been no efforts at the federal, state, county or provider level to identify or promulgate mental health planning best practices for planners to follow and planning tools for planners to use. We will discuss such best practices and tools in subsequent posts.

References
No Authors Listed (1979). “Mental health leaders angered by deferral of $30.5 million in funding for new CMHCS.” Hosp Community Psychiatry 30(6): 422-423.
Bachrach, L. L. (1979). “Planning Mental Health Services for Chronic Patients.” Hosp Community Psychiatry 30(6): 387-393.
Bell, N. and D. Shern(2002). State Mental Health Commissions: Recommendations for Change and Future Directions, NationalTechnicalAssistance Center for State Mental Health Planning (NTAC)
National Association of State Mental Health Program Directors (NASMHPD).
Bloche, M. G. and F. Cournos(1990). “Mental HealthPolicy for the 1990s: Tinkering in the Interstices.” Journal of Health Politics Policy and Law 15(2): 387-411.
Cochrane, A. L. (1972). Effectiveness and efficiency: random reflections on health services. London ,, Nuffield Provincial Hospitals Trust.
Drake, R. E. and G. R. Bond (2008). “Information Technology and Evidence-based Practices: A Commentary on “Evidence-Based Implementation Strategies: Results of a Qualitative Study”.” Community Ment Health J 44(3): 225-226.
Drake, R. E., H. E. Goldman, et al. (2001). “Implementing evidence-based practices in routine mental health service settings.” Psychiatric Services 52(2): 179-182.
Fisher, D. B., J. Chamberlin, et al. (2005). The Role of Mental HealthConsumers in Leading the Recovery Transformation of the Mental HealthSystem. Universal healthcare: Readings for mental health professionals. Reno, NV, US, Context Press: 219-242.
Frank, R. G., T. G. McGuire, et al. (1999). “The value of mental health care at the system level: the case of treating depression.” Health Aff 18(5): 71-88.
Goldman, H. H. and J. P. Morrissey (1985). “The alchemy of mental health policy: homelessness and the fourth cycle of reform.” Am J Public Health 75(7): 727-731.
Grob, G. N. and H. H. Goldman (2006). The dilemma of federal mental healthpolicy : radical reform or incremental change? New Brunswick, N.J., Rutgers University Press.
Gunnison, R. B. (1966). What Tammany Hall Can Tell Us About Sacramento. San Francisco Chronicle.
Koyanagi, C. (2007). Learning From History::Deinstitutionalization of People withMental Illness As Precursor to Long-Term Care Reform. Kaiser Commission on Medicaid and the Uninsured. Washington DC.
Koyanagi, C. and H. H. Goldman (1991). “The Quiet Success of the National Plan for the Chronically Mentally Ill.” Hosp Community Psychiatry 42(9): 899-905.
Minnesota Psychiatric Sociey. (2004). “Minnesota Mental Health System: Demand, Capacity and Cost.” Retrieved May 3, 2010, from http://www.mnpsychsoc.org/MH2004.pdf.
National Alliance to End Homelessness (2000). A Plan, Not a Dream: How to End Homelessness in Ten Years. Washington, DC.
National Council on Disability (2003). Olmstead: Reclaiming Institutional Lives. Washington DC.
New Freedom Commission (2003). Achieving the Promise: Transforming Mental Health Care in America.
North Carolina Department of Health and Human Services (2007). Transformation of NorhCarolina’s System of Services for Mental Health, Developmental Disabilities and Substance Abuse: 2007-2010: 1-41.
Osborne, D. (1990). Laboratories of democracy. Boston, Mass., Harvard Business School Press.
Palmer, A. (No Date). “20thCentury History of the Treatment of Mental Illness: A Review.” Retrieved May 3, 2010, from http://web.archive.org/web/20040710022826/http:/www.mentalhealthworld.org/29ap.html.
Pirkis, J., M. Harris, et al. (2007). “InternationalPlanning Directions for Provision of Mental HealthServices.” Administration and Policy in Mental Healthand Mental Health Services Research 34(4): 377-387.
SAMHSA’s NationalMental HealthInformation Center: Center for Mental HealthServices. (2010). “Community Mental Health Services Block Grant Program.” Retrieved 6/01/2010, from http://mentalhealth.samhsa.gov/publications/allpubs/KEN95-0022/#what.
Substance Abuse and Mental Health Services Administration (2010). “Mental Health Transformation State Incentive Grant Program.” from http://mentalhealth.samhsa.gov/cmhs/communitysupport/mentalhealth/default.asp.
Sharfstein, S. S. (2000). “Whatever happened to community mental health?” Psychiatr Serv 51(5): 616-620.
United States. Dept. of Health and Human Services. Steering Committee on the Chronically Mentally Ill. (1980). Toward a national plan for the chronically mentally ill : report to the Secretary. Washington, D.C.?, U.S. Dept. of Health and Human Services, Public Health Service.

VN:F [1.6.2_892]
Rating: 10.0/10 (2 votes cast)

I recently presented on this topic to the 2010 National Mental Health Research to Policy Forum of The International Society for CNS Clinical Trials and Methodology.  A version of the presentation is in the PowerPoint  reachable by the link below (note: may take a while to open).

One  topic was how clinical trials data could be best represented to support mental health systems planning and systems comparative effectiveness research. 

Another was how the culture that has grown up around mental health planning needs to be changed to develop mental health plans that are effective and safe, not just “aspirational.”  The details of what I mean by aspirational, effective, safe and systems comparative effectiveness research are in the PowerPoint.

To see the presentation, click on the link below.  The PowerPoint begins with a scene from an EDS video, which is an early post to the blog, for that scroll down in the blog and go to “older posts. ”  Here’s the link to the PowerPoint:

Hitchhiker’s Guide to MH System Reform Publication Version

Virtually,

Steve

VN:F [1.6.2_892]
Rating: 0.0/10 (0 votes cast)

Planning is often analogized to creating a map. For example see: The Campaign for Mental Health Reform’s EMERGENCY RESPONSE: A Roadmap for Federal Action on America’s Mental Health Crisis (2005)

(http://www.mhreform.org/Portals/0/1.3_EmergencyResponseReport.pdf)

I am writing an article or book chapter tentatively titled Planning Reforms and Reforming Planning: Technical Planning for Mental Health Reform. So I have been thinking and reading a lot about planning lately.  I will describe what I mean by technical planning below.

But before I do this is a disclaimer:  Technical planning (planning by the numbers) is what I do.  So I and colleagues have developed planning tools to do this – specifically computer and web-implemented simulation and other models.  This has two implications: one is I am always trying to get people to use these models, and that is one, but not the main, reason for this post.  The second implication is I have biases I recognize and probably ones I don’t.  I’m not going to go into all these in this post, but if anyone wants to comment on these, I will try to respond.

Compasses

My analysis at this point is that many of the plans and commission reports (which are treated as sort of plans) are not roadmaps. They are more like compasses; they indicate the directions change might or should take, but they do little to actually supply details of how to get from where we are to where we would like to go.  Another way of thinking about these “plans” is that they embody the visionary or values component of planning and as that they are an important component of planning. 

Visionary planning has two main purposes.  One is to set directions.  The other is to do what Walter Lippman in 1922 called “the enlistment of interest” (Lippmann, W. (2004). Public opinion. Mineola, N.Y., Dover Publications).  Both are necessary advocacy functions and provide energy to technical planning.

In fact, even when you look at reports actually labeled maps like the Campaign for Mental Health Reform’s they aren’t maps either, they are more llike compasses; designed more to set directions and enlist stakeholder interest, not to guide the details of action.

MapQuest

MapQuest is an example of a map: one that uses modern technology that gives us directions about how to get from where we are to where we want to go.  Not only that, it gives us mileage, fuel estimates, fuel cost estimates, traffic reports and more.  On the odd chance you have never used MapQuest, the link below is from Harvard College to my favorite Chinese restaurant, Chang Sho!  This is close to what I mean by technical planning.  But with PC and Internet technology we can do much better.

http://www.mapquest.com/mq/5-JUL6hOVF

The GPS: Technical Planning

So what is better: consider the GPS.  It gives you everything MapQuest does, plus real time information on where you are, your speed, current traffic etc.  If you make a wrong turn or decide to take a side road it didn’t plan for you, it simply recalculates directions for you, although the “Recalculating” message can be annoying.

We should be using planning technology in mental health that is more like a GPS.  By this, I mean simulation and other models that request us to specify data about and can give us projections of numbers of persons to be served, services needed, costs of services, reimbursements, positive and negative outcomes to be expected, etc.  This information should be easy to enter and change as traffic is encountered.  And it should give us specific milestones to track.  This is not impossible to do.  Models like this are built by operations researchers for industry all the time.  And we, and others,  have built a few for mental health.  But I’ll save that for another post.  But although some of these models have been around for quite a while (see Hargreaves, 1986)

http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=3718171

 these models are not widely used at the Federal, State and county levels for mental health planning (e.g., Block Grant Planning).

Technical Planning and Incrementalism

Grob and Goldman (2006) contrast what they call “radical reform” with “incremental change.” 

Grob, G. N. and H. H. Goldman (2006). The dilemma of federal mental health policy : radical reform or incremental change? New Brunswick, N.J., Rutgers University Press.

I think radical reform is compass guided reform.  I prefer to think of it as visionary planning.  Visionary planning gets people all fired up, but it doesn’t realistically portray what is needed to stoke the fire and often results in failure and disappointment, because, in ways that models easily predict, in the end dollar and human resources don’t match needs.  This can discredit the vision, its advocates or both, making it more difficult to gain support for subsequent attempts at reform.

Incremental change is opportunistic change.  It is sort of like I used to drive before MapQuest and my GPS.  I knew where I was going and if I was stuck in traffic and saw what looked like a less crowded road I took it, often getting lost, and going through the whole “men don’t ask directions” thing.  The reality is that sometimes you make progress by being opportunistic and sometimes you lose your way.  Also sometimes you sacrifice long term gains for short term ones.  A great discussion of this can be found in Robert Mayer’s book: Policy and Program Planning: A Developmental Perspective (Mayer, R. R. (1985). Policy and program planning : a developmental perspective. Englewood Cliffs, N.J., Prentice-Hall).

Technical planning bridges the gap between visionary planning and incremental planning.  It addresses all the details of planning how to get from here to there (including considering options).  But because it is computer-implemented it allows for considering incremental changes as they occur and it enables planners to “Recalculate” and see how opportunistic changes might affect the achievement of long term goals.

Technical Planning and Politics 

Some people might think I am naive about what happens with plans when they meet politics.  You’ve probably heard of Murphy’s Military Law #2: No battle plan ever survives contact with the enemy.

My GPS experience taught me something about this.   When I first got my GPS I thought it would allow my wife to forgo worrying about whether I knew where I was going (and whether I was even attending to that) and eliminate arguements about what routes to take.  To make a long story short: my wife began to argue with the GPS.  For certain destinations she used different algorithms than the GPS.   And as far as I can tell, in areas my wife was familiar with, her routes are better.  So we had to learn and negotiate about when to use the GPS and when not to; and over time we came up with mixed strategies keyed both to our sense of what the GPS was good for and what not, and also our preferences for giving and following directions.  My experience is that technical planning requires this and equally important makes political negotiation and deliberation (see my post on deliberative planning) more possible since options can be explored and then debated, rather than debated in the absence of data on possible costs and outcomes.

Anyway, running out of steam on this, but may write more later and will certain elaborat in Planning Reforms and Reforming Planning.

VN:F [1.6.2_892]
Rating: 0.0/10 (0 votes cast)

Outcomes produced by Markov Transition Probability Analysis (MTPA) are particularly useful for planning.  Mental health planners should urge evaluators and researchers to use MTPA.   I discuss why below.

The references for this post are very interesting, but unfortunately, too many for me to attach links.  I have put in links to a two of the best: Hargreaves (1986) and James et al. (2006).

I am looking for articles relevant to MTPA and data sets people are willing to share.  If you know of published or unpublished articles, I would appreciate references or author contact information.  Please pass along information via comments.  If you know of any data sets I would also appreciate hearing about these via comments.

Markov Transition Probability Analysis

What is Markov Transition Probability Analysis
There are many approaches to modeling systems planning and evaluation, including different versions of MTPA (Levin and Roberts 1976; Pettiti 2000; Catalano, McConnell et al. 2003; Willan and Briggs 2006).  Discussing all of these is beyond the scope of this post, but readers are encouraged to consult the overviews cited above.

I am interested in one of the simpler forms of MTPA, MTPA for cohort, first order, and stationary Markov transition probabilities. These terms are explained below.

The basic concepts of MTPA for application in health systems planning and evaluation are discussed in a number of places. The citations following are only some examples (Hargreaves 1986; Sonnenberg and Beck 1993; Briggs and Sculpher 1998; Bloom and Bloom 1999; Patten and Lee 2004; Bala and Mauskopf 2006; Brennan, Chick et al. 2006; Sonnenberg 2009).

MTPA applied to health is based on the idea that at any point in some time period (e.g., day, month, year) a person occupies one of some number of outcome states defined by a condition of interest.  In my own work I use a functional level scale I developed, the Resource Associated Functional Level Scale (RAFLS).  If you are interested in the RAFLS please contact me via comments.

A second MTPA idea is that from one time period to the next the person may transition from the state that he or she is in to some other state. MTPA states can be ones from which persons can transition, or they can be ones from which persons can not exit (absorbing states). Death is an example of an absorbing state.

Cohort models describe outcomes for proportions of persons in groups in contrast to individual (Monte Carlo) models (Briggs and Sculpher 1998).

First Orderness

First-order MTPA posits that beyond its present state, a cohort’s history (e.g., diagnostic and/or service use) is not relevant to its next transitions (Briggs and Sculpher 1998; Brennan, Chick et al. 2006). This will seem counter-intuitive to most persons knowledgeable about mental health disorders and system. When history is known or believed to be determinative of transitions, my recommendation is to implement different first-order MTPAs for different subgroups defined by the historical variable (Miley, Lively et al. 1978), an approach providing the most useful information for planning and evaluation.

Stationarity

Stationary transition probabilities are ones that change relatively little over time (Drachman 1981; Hargreaves 1986).  In theory, the more stationary tranition probabilities are, the more accurate they will be for any time period, and the more useful they are for planning.  However, there is some evidence that lack of stationarity does not cause substantial inaccuracy (Heard 1981).  Additionally, if a lack of stationarity appears extreme, this can be addressed by implementing different MTPAs for different time periods. This strategy is akin to addressing violations of first-orderness by implementing MTPAs for different subgroups.

At any point in time, a cohort of individuals can be described by the proportions of persons occupying the various outcome states.  When MTPA is applied to mental health, outcomes are distributions of persons by mental health related outcome  states (e.g. functional levels).  Below is a functional level transition matrix developed for one system we studied.  These transition probabilitiies are presented for illustrative purposes only and I don’t advise using them for other systems.

Example of a Markov Transition Probability Matrix

  Destinations
Initial Functional Level 1 2 3 4 5 6 7  Syst Indep Disap-pear Death Total
1 Extreme Risk .721 .136 .058 .022 .008 .006 .000 .043 .006 1.000
2 Moderate Risk .036 .570 .224 .104 .027 .000 .000 .036 .004 1.000
3 Lacks ADLs .006 .067 .689 .177 .021 .001 .000 .036 .003 1.000
4 Lacks Commun Skills .017 .019 .069 .732 .111 .013 .000 .037 .003 1.000
5 Vulnerable to every day stress .013 .007 .015 .076 .758 .091 .000 .038 .002 1.000
6 Stable-needs maintenance/wants treat .000 .008 .000 .008 .092 .853 .003 .036 .001 1.000

Uses of MTPA

Health services scientists have shown that MTPA outcome states can be used in needs assessment (Hargreaves 1986), cost-effective analysis (Sonnenberg and Beck 1993; Briggs and Sculpher 1998; Bala and Mauskopf 2006; Brennan, Chick et al. 2006), performance measurement (Miley, Lively et al. 1978), estimating incidence, prevalence and course (Patten and Lee 2004; Patten 2005; Bala and Mauskopf 2006), and for system services planning (including both types and amounts) and resource allocation (Bala and Mauskopf 2006; James, Sugar et al. 2006; Leff, Hughes et al. in press).  I have found MTPA  particularly useful for planning and costing service systems to inform law suits involving the Federal Government and States and counties (Leff, Hughes et al., in press).

Ways in which MTPA Is Superior to Single Number Measures of Outcomes or Prognosis

There are at least three important advantages MTPA has over single number (e.g., means) measures of outcome.

  • First single number outcomes, like means, promote the misconception that if a treatment or service is effective on average every person receiving the treatment or service is characterized by the mean. In fact, even for interventions effective on average, some people improve, while others remain the same, and still others actually become worse. MTPA captures this more detailed view of outcomes, useful for treatment and program planning (James, Sugar et al. 2006; Kent and Hayward 2009). Given information on who did not respond to interventions or responded negatively, researchers can investigate the reasons for this and what treatments and programs might be effective for these subgroups.
  •  Another advantage of MTPA is that any type of weighting can be associated with health states, including costs, revenues, utilities (e.g., QALYS), performance weightings (Miley, Lively et al. 1978), hand preferences (Lenert, Sturley et al. 2004).
  • A third advantage of MTPA is that transition probabilities, if they pass certain statistical tests, can be used  to project for any number of time periods outcomes for service plan options.  We also can project any service and utility measures associated with these outcomes (Briggs and Sculpher 1998; James, Sugar et al. 2006).  These projections then can be used in selecting service options.   This is the ”planning for outcomes” part. 

References
Bala, M. V. and J. A. Mauskopf (2006). “Optimal Assignment of Treatments to Health States Using a Markov Decision Model: An Introduction to Basic Concepts.” PharmacoEconomics 24(4): 345-354.
Bloom, L. A. and B. S. Bloom (1999). “Decision analytic modeling in health care decision making. Oversimplifying a complex world?” Int J Technol Assess Health Care 15(2): 332-339.
Brennan, A., S. E. Chick, et al. (2006). “A taxonomy of model structures for economic evaluation of health technologies.” Health Econ 15(12): 1295-1310.
Briggs, A. and M. Sculpher (1998). “An introduction to Markov modelling for economic evaluation.” PharmacoEconomics 13(4): 397-409.
Catalano, R., W. McConnell, et al. (2003). “Psychiatric emergency services and the system of care.” Psychiatr Serv 54(3): 351-355.
Drachman, D. (1981). “A residential continuum for the chronically mentally ill: a Markov probability model.” Eval Health Prof 4(1): 93-104.
Hargreaves, W. A. (1986). “Theory of psychiatric treatment systems. An approach.” Arch Gen Psychiatry 43(7): 701-705.

http://archpsyc.ama-assn.org/cgi/content/abstract/43/7/701

Heard, E. (1981). “A Simulation Investigation of the
Markov Stationarity Assumption.” Journal of Experiential Learning and Simulation 2: 239-251.
James, G. M., C. A. Sugar, et al. (2006). “A comparison of outcomes among patients with schizophrenia in two mental health systems: A health state approach.” Schizophrenia Research 86(1): 309-320.

http://www.ncbi.nlm.nih.gov/pubmed/16806839

Kent, D. and R. Hayward (2009). “When Averages Hide Individual Differences in Clinical Trials.” American Scientist 95(January-February): 60-68.
Leff, H. S., D. R. Hughes, et al. (in press). A Mental Health Allocation and Planning Simulation Model: A Mental Health Planner’s Perspective. Handbook of Healthcare Delivery Systems. Y. Yih, Taylor & Francis
Lenert, L. A., A. P. Sturley, et al. (2004). “Public preferences for health states with schizophrenia and a mapping function to estimate utilities from positive and negative symptom scale scores.” Schizophr Res 71(1): 155-165.
Levin, G. and E. B. Roberts (1976). The dynamics of human service delivery. Cambridge, Mass., Ballinger Pub. Co.
Miley, A. D., B. L. Lively, et al. (1978). “An index of mental health system performance.” Evaluation Quarterly 2(1): 119-126.
Patten, S. B. (2005). “Modelling major depression epidemiology and assessing the impact of antidepressants on population health.” International Review of Psychiatry 17(3): 205-211.
Patten, S. B. and R. C. Lee (2004). “Epidemiological theory, decision theory and mental health services research.” Social Psychiatry and Psychiatric Epidemiology 39(11): 893-898.
Perry, J. C., P. W. Lavori, et al. (1987). “A Markov model for predicting levels of psychiatric service use in borderline and antisocial personality disorders and bipolar type II affective disorder.” Journal of Psychiatric Research 21(3): 215-232.
Pettiti, D. B. (2000). Meta-Analysis, Decision Analysis, and Cost-Effectiveness Analysis: Methods for Quantitative Synthesis in Medicine Oxford, Oxford Univeristy Press Inc.
Sonnenberg, A. (2009). “Transposed Markov matrix as a new decision tool of how to choose among competing investment options in academic medicine.” Computational & Mathematical Methods in Medicine 10(1): 1-7.
Sonnenberg, F. A. and J. R. Beck (1993). “Markov Models in Medical Decision Making: A Practical Guide.” Med Decis Making 13(4): 322-338.
Willan, A. R. and A. H. Briggs (2006). Statistical Analysis of Cost-Effectiveness Data West Sussex, John Wiley & Sons.

VN:F [1.6.2_892]
Rating: 0.0/10 (0 votes cast)
  • Health States: Outcomes for Planning and Planning for
    Outcomes
  • Markov Transition Probabilities  in Mental Health Planning
  • Focus Groups versus Individual Interviews for Mental
    Health Planning
  • And more…I hope
VN:F [1.6.2_892]
Rating: 0.0/10 (0 votes cast)

It has become a habit and often a requirement that planning processes include town halls, forums, and other forms of public meetings for soliciting public, often termed “stakeholder”, input for public mental health plans.  Recently, for example, Massachusetts held five public hearings throughout the state at which three hundred people ”voiced their observations, experiences, and concerns” about plans for the State’s adult inpatient psychiatry system (James T. Brett and Marylou Sudders Boston Globe, 7/28/2009).

Another way of gaining input for mental health planning is to conduct stakeholder key informant interviews and focus groups as parts of “needs assessments”(Rossi, Lipsey et al. 2004).

These efforts represent laudable attempts to enable public participation in mental health planning. However, they often serve primarily as platforms for advocacy rather than as means for discussing decisions and plans that solve acknowledged problems. As such they typically fail to obtain new information and in the long run may lead to more stakeholder unhappiness when the information provided is not reflected in eventual decisions and plans. The “town halls” recently held on the topic of healthcare reform are extreme examples of how such meetings can become more expressive than informative or deliberative (see column by Fishkin, below).

Now that we have established the principle that the public deserves to be heard and participate in mental health planning, it’s time to develop ways that this participation can be more instrumental, satisfying and meaningful. This is especially true given the increasing power of the internet to enable virtual public hearings. We could take steps to accomplishing this by:

  • Making sure that ideas are solicited from representatives of all stakeholder groups, not just those organized and resourced enough to come to meetings, prominent enough to be interviewed or assertive enough to be heard in meetings. This requires both sampling that is representative and use of group facilitation techniques that counteract social psychological phenomena that cause some opinions not to be voiced or heard (Sunstein 2006; Sunstein 2007).
  • Providing information in advance of meetings to persons whose opinions are being sought to allow them to consider not just their own needs and preferences, but also the needs and preferences of others and the resource constraints that often mean that not every need and preference can be met.
  • Facilitating exercises of some sort that involve iteratively prioritizing needs and preferences and considering how these must be adjusted and compromised to take into account resource constraints. In subsequent postings I will discuss mental health planning simulations as tools that can be used for this purpose.

Recently, I’ve been impressed by an approach to involving the public in decision making known as “deliberative democracy”. This approach includes the ideas above and puts the discussion in the context of the United States’ political history and traditions.  I don’t know if all public meetings and needs assessments ought to follow this model, but I think many should and almost all should incorporate features of the model that encourage enlightened deliberation.  Research papers from the Center for Deliberative Democracy can be found at:
http://cdd.stanford.edu/research/

Three particularly relevant papers are:

Fishkin (2009) Town Halls by Invitation.  This New York Times column is a perspective on the healthcare reform town halls from a deliberative democracy point of view.

http://www.nytimes.com/2009/08/16/opinion/16fishkin.html?scp=1&sq=town+hall+deliberative&st=nyt

Fishkin (2000) Virtual Democratic Possibilities: Prospects for Internet Democracy1
http://cdd.stanford.edu/research/papers/2000/brazil_paper.pdf

Fishkin and Luskin, (2005) Experimenting with s Democratic Ideal: Deliberative Polling and Public Opinion
http://cdd.stanford.edu/research/papers/2005/acta-politica.pdf

I was thinking about this recently when I visited the Norman Rockwell Museum in Stockbridge Massachusetts (http://www.nrm.org/).  (I recommend a visit.  Rockwell is an artist, not just an illustrator, who movingly captured important aspects of American life.)  I was struck by two Rockwell paintings.  They seemed to me to depict two aspects of deliberative planning:

Freedom of Speech (Note that participants have data in pocket or hand)

The Right to Know


References

Rossi, P. H., M. W. Lipsey, et al. (2004). Evaluation : a systematic approach. Thousand Oaks, CA, Sage.

http://www.amazon.com/Evaluation-Systematic-Dr-Peter-Rossi/dp/0761908943/ref=sr_1_1?ie=UTF8&s=books&qid=1254604141&sr=1-1

Sunstein, C. R. (2006). Infotopia : how many minds produce knowledge. New York, Oxford University Press.

http://www.amazon.com/Infotopia-Many-Minds-Produce-Knowledge/dp/0195340671/ref=sr_1_5?ie=UTF8&s=books&qid=1254604294&sr=1-5

Sunstein, C. R. (2007). Republic.com 2.0. Princeton, Princeton University Press.

http://www.amazon.com/Republic-com-2-0-Cass-R-Sunstein/dp/0691143285/ref=sr_1_6?ie=UTF8&s=books&qid=1254604294&sr=1-6

VN:F [1.6.2_892]
Rating: 0.0/10 (0 votes cast)

What This Blog Is About

This blog is about how we might  improve mental health systems planning using web-implemented tools like :

  • Databases for detailing the contents of plans
  • Simulations for evaluating planning options, and
  • Media for communicating about plans, sharing expertise, and building a supportive community for mental health planners

I will provide references to and possibly review articles, reports and other resources I find interesting or important  for mental health planning.  The blog will also link to  trial versions of applications like databases and simulations.

Post your own messages as the spirit moves you.  I’ll be moderating the blog to keep the topic on mental health planning, particularly quantitative planning, and to see that things don’t get personal.  If you’re not sure a message is right for the blog, I encourage you to send it along.

I hope you find Mental Health Planning by the Numbers Useful and the basis of a supportive community.

Virtually,

Steve Leff,  Ph.D.

Organizational Afffiliations

**Senior Vice President,
Human Services Research Institute

**Associate Professor
Harvard Medical School
Department of Psychiatry
At The Cambridge Health Alliance

**The opinions expressed on Mental Health Planning by the Numbers are Steve Leff’s or commenters’, and in no way represent positions of either The Human Services Research Institute or the Harvard Medical School Department of Psychiatry at The Cambridge Health Alliance

I think all mental health planners have cartoons and jokes they use with planning groups.  I’ll be sharing some here in different ways as I understand the copywrite issues.  Below is a link to one of my favorites about the need for evidence-based mental health planning: “Then a miracle occurs…” 

Too often our plans don’t make clear what outcomes we expect from planned services and how these outcomes will change service needs.  One result is that we project tomorrow’s needs based on today’s services, although our plans call for changing these services.  If we assume that changes in services improve outcomes, thereby reducing the need for more intensive services, then planning without taking outcomes into account may mean we are planning for more services and expenditures than should be needed. 

There are more good science cartoons here

http://www.sciencecartoonsplus.com/pages/gallery.php

Virtually,

Steve

VN:F [1.6.2_892]
Rating: 0.0/10 (0 votes cast)

I have been reviewing mental materials on mental health commissions and planning.  In particular, I’ve been thinking about why we’re not making more progress given how many commissions and plans there have been (more on this later…).   For an interesting report on mental health commissions prepared for the National Association of State Mental Health Directors by Nancy Bell and Dave Shern , see:

http://www.nasmhpd.org/general_files/publications/ntac_pubs/reports/statementalhealthcommissions.pdf

Then I remembered this joke. 

A resigning public official, overwhelmed by crises, was briefing his successor. He said, “I’m leaving you three envelopes. When you have your first crisis, open the one numbered one. For your second, the one numbered two, and for your third the one numbered three.” When she encountered her first crisis, the new official opened the first envelope. It said: Blame Me. So the new official blamed the previous one. When she encountered her second crisis she opened envelope number two. It said: Appoint a Commission. So the new official did that. Then she encountered her third crisis and she opened envelope number three. It said: Prepare Three Envelopes.

Virtually,

Steve

VN:F [1.6.2_892]
Rating: 0.0/10 (0 votes cast)

In a previous post I indicated I really believe mental health plans should be presented in web implemented searchable databases.  Transformation Tracker Explorer is an example of such a plan database

Transformation Tracker Explorer

Transformation Tracker Explorer (http://mhtsigdata.samhsa.gov) is an innovative new website recently launched to let the public know about the activities that States are pursuing as a result of the SAMHSA Center for Mental Health Services (CMHS) Mental Health Transformation State Incentive Grants (MHT SIG).  Persons interested in mental health change are encouraged to explore the site, and to share it with others who might find it interesting or useful.   Transformation Tracker Explorer was developed by the cross-site evaluation team.  The SAMHSA CMHS Project Officer for the cross-site evaluation is Dr. Crystal Blyler (CBLYLER@samhsa.gov)

Members of the public can search the site to find transformative ideas about specific topical areas (such as rural mental health care, consumer employment services, linking with the criminal justice system, etc.) that they can implement or advocate for locally.  Through the site, stakeholders can also engage in discussion about specific transformational activities.

The information on this site links directly to the data that MHT SIG States provide to SAMHSA as part of the program’s cross-site evaluation.  The underlying database was developed to record all the infrastructure activities that grantees have planned to make throughout the course of the grant.  Because the site links directly to the cross-site evaluation database, the information on Explorer will be continuously updated as grantees enter new data.

Transformation Tracker Explorer1 may be the first ever web site to provide stakeholders with a view into a U.S. federal mental health grant program while it is on-going along with an opportunity to rate and comment on State activities.  As such it allows for “direct to stakeholder” evaluation for use in the program’s summative evaluation.  We encourage you to leave comments and ratings.

1Transformation Tracker Explorer is powered by HSRI Transformation Manager.  We hope to have a basic version of Transformation Manager available for trial use on this blog in the future.  For additional information contact: Ben Cichocki at Human Services Research Institute: (bcichocki@hsri.org)

Virtually,

Steve

VN:F [1.6.2_892]
Rating: 0.0/10 (0 votes cast)

Older Posts »