The Mental Health Services Act – Then and Now

For a variety of reasons, I am being asked much more than ever about the thinking behind various provisions of the mental health services act when we are writing it in 2003 and how that relates to some of the challenges we now have.

Not surprisingly, most of the challenges we now are seeing involved issues we anticipated when we wrote the Act.  We have posted all of the preliminary drafts of the MHSA on our website in case anyone is curious about how the language evolved from the first draft in May 2003 to the final language in September.

The Act still looks like it correctly identified what was needed to fulfill our goals.  But many parts of it have not been implemented the way it was envisioned and a lot has changed from what we knew in 2003 when it was written. In 2006 I had a different way of viewing how would be implemented and significantly underestimated how challenging it would be for state agencies to regularly update regulations which I thought could be an ongoing process with updates every three years. I also significantly underestimated the delays and challenges in developing outcome data.

In this blog I will speak to some of the issues and some of the solutions.  Mostly what is required is a recognition that we have not revisited the guidelines (which led to regulations), all of which were developed before the applicable parts of the act had been implemented.

Now that we have had many years of implementing each part of the act is time to revisit the guidance (leading eventually to updated regulations) through a series of separate workgroups, that can move us more firmly in the direction that I think we all know is where we need to go.

Key recommendations (all of which have been previously stated in different contexts):

Separate the children’s system of care from the adults and older adult system of care – dividing the funding now known as community service and supports into the two separate system of care programs and completely re-write the guidelines for each.

  • For children, focus on the uses of funds for services that are not covered by EPSDT or AB 114 to support families.
  • For adults and older adults clarify the different levels of need within full-service partnerships, the relationship to housing and homeless outreach and best cost-effective uses of funds not expended for full-service partnerships.
  • For both ensure that there are performance outcome systems comparing all providers (not just counties) to identify the best approaches (for each of the many different types of problems and populations being served) that others should replicate. For both identify how to incorporate the alcohol and drug organized delivery system.

Identify core prevention and early intervention strategies most likely to make early identification a norm (schools, primary care, workplace, technology, youth centers, and others), using innovations funds to develop the models for replication and eventually steering the majority of prevention and early intervention (PEI) funds to only the proven most effective models. Ensure that each model addresses all behavioral health issues and creates warm handoffs to appropriate service providers.

Rewrite guidelines and regulations for the process to develop local plans. Ensure useful fiscal transparency for stakeholders, and meaningful interactive process after a draft county spending plan has been developed, focusing on quality rather than quantity in measuring stakeholder participation. Create local steering committees for prevention and early intervention and performance evaluation then bring in the diversity of outside interests, similar to the OAC.

Develop a long-range workforce strategy that provides guidance to County plans and creates financial participation for employers of behavioral health professionals outside of the county systems.

Develop capital facilities strategies to take SB 82 and children’s crisis funding to scale so that no one is taken to a jail or hospital simply because there were no alternative crisis care options.

In anticipation of the election of a new governor, develop clear roles and responsibilities for the OAC, DHCS, planning Council and other agencies that ensures there is a single specific place in state government that has clear authority, commitment and resources to address each major issue.

Analysis that led to these recommendations follows.


The primary funding for children’s services come from the two main entitlements: EPSDT and AB 3632.   The children system of care is a broader family oriented care coordination model that allows for payment for services that might not be covered under either of those entitlement programs.  The children system of care part of the act was expected to be used to provide supplemental funding to eliminate those gaps so the children and families received “whatever it takes” to maximize success.

Nothing has gone according to plan.  The two entitlement programs remain in place but the state has “realigned” them to counties and schools – and there is ample evidence that fewer children are being served under EPSDT (penetration rate went from 5.7% to 4.8% in first two years) and many schools are providing less comprehensive services to children with serious emotional disturbances and counties have been doing under AB 3632.

The merging of the children’s and adult systems of care under community services and supports led counties to believe that full-service partnerships for children were to be freestanding separate programs rather than limited supplemental services to children already being served through the entitlement programs.  I have always had trouble understanding who the children were who could meet the eligibility requirements of serious emotional disturbance and not qualify for either of the two entitlements.  It seems better to start all over and rewrite the guidelines focusing on the types of services and circumstances under which a family night need something not covered by EPSDT or AB 114 or other Medi-Cal or insurance.

The EPSDT performance outcome system is intended to create the outcome measures necessary to determine the extent to which programs are successfully serving children with serious emotional disturbances.  The children system of care funded services need to be included in this evaluation and incorporate all MHSA funds.

Similar outcome measures need to be collected for AB 114 special education funded mental health services for children with serious emotional disturbances so that we can learn what is working and where and why.


The Act did not create the adult system of care. It existed for many years prior and was the funding source for the AB 34 and AB 2034 programs.  Its definition of comprehensive “whatever it takes” services did not require that everyone had to stay at a very high level of service in order to continue to be eligible for services.  As people progressed along the milestones of recovery we expected lower levels of care that would still be part of the services for people with severe and disabling mental illnesses unless and until the recovery led to full-time employment in which case we expected they would leave the system of care and presumably Medi-Cal.

Somehow the regulations for the Act got interpreted to be much more rigid and narrow than was intended. That should be corrected. Similarly it was always clear in the Act and previous law that housing was one of the eligible uses of the funds. That also needs to be corrected – and there is legislation intended to do so, which shouldn’t have been necessary.

While we have an outcome measurement system that collects information on full-service partnerships we still don’t have the type of reports that we had for AB 2034 and always expected to have under the Act. These would provide real-time quarterly comparisons of each program by each provider measuring progress in achieving the primary functional outcomes. The data has always been collected from providers to produce these reports, but in most counties the providers have never seen any reports, and none have been created or collected at the state level except comparing counties, not providers.

This prevents the reports from identifying the providers who are achieving the best outcomes so that everyone else can learn what they are doing better or differently and learn to replicate it.

Another oversight in the implementation of the Act has been a failure to collect information on the use of community service and support funds that do not go directly to full-service partnership programs.

We know some is used for administration and planning (including some stakeholder processes), some is used for outreach and engagement, and some is used for services other than full-service partnerships.

Exactly how much is used for each purpose in each county should be information readily available for each year.  It is likely that some of the more common uses lend themselves to comparative performance measurement.

An update of the guidelines and reporting systems should allow the development of this information.


In developing the scope of the Act, voters were clear that they only wanted to pay for people with severe mental illnesses, and for prevention and early intervention. They did not support funding for less severe mental illness or for substance use disorders except when those conditions needed to be treated incidental to a severe and disabling mental illness or the prevention of a severe mental illness.

From the outset of AB 34 (which became full-service partnerships) it was recognized that many of the people to be served would have co-occurring substance use disorders and that the funding was intended to pay for those services to the extent that other funding was not available. This appears to have worked as intended, but we don’t have good data on the extensiveness of the use of these funds for that purpose and how consistently this is being done.  With the development of the county organized delivery system there should be a better way to more broadly get federal financial participation for the services.

The application of prevention and early intervention for people with substance use disorders has not been identified as a separate priority and as best as we can tell none of the funds have been expended to support substance use disorder treatment.

As we implement core prevention and early intervention strategies that I describe below, they must all be for comprehensive behavioral health including substance use disorders.


the housing bond measure approved by the Legislature in 2016 was viewed by many as a major departure from the original purposes of Proposition 63.  That is not correct.  As we were writing the ballot measure, we recognized that a significant portion of the target population was homeless and that using MHSA funds to leverage other housing funds through a bond measure was anticipated to be something we would do after the measure was passed.  We did not highlight it in the text of the measure because we feared that it could create a NIMBY backlash, but we did have legal opinions telling us that it would be a proper use of the funds.

Like any major new effort, the “No Place Like Home” housing program will present challenges that will have to be worked out over the many years of its implementation – that depend upon many variables such as:

  • transitional housing for those for whom permanent supportive housing is not the right approach
  • unknown amounts of housing that can be leveraged from other funding sources that reduces the cost per unit from MHS a funds
  • unknown rates of growth of MHSA funds that will be needed to ensure services to those newly housed to not displace services to others
  • uncertainty of other funding sources

County MHSA expenditure plans need to account for all of the housing needs of the target population including those which cannot be met through No Place Like Home – and describe strategies to address those needs.  An update of the system of care guidelines and regulations should address these issues.


the Act set aside $450 million for this program to remedy the shortage of qualified individuals to provide services. That money is nearly all spent, but the shortages are not gone – and if anything they are greater now than when the Act was written.

There is a continued need to do what the Act requires, which is to have a state plan every five years that relies upon each county doing a needs assessment.

However, it is not just the public mental health system that has benefited from the funds already set aside, and which is impacted by the shortages that still exist.

Schools, jails, prisons, and commercial health plans all employ individuals and compete with the public mental health system to hire and retain qualified individuals.

With the state set-aside funding ending in 2017-18 there is a need for a new state plan to provide guidance going forward.  That guidance should address whether there is a need to have a reinstated set-aside of funds from County community services and supports allocations and how to create financial partnerships with other types of providers and higher education to address the broader needs for behavioral health workforce beyond MHSA funded services.


We set aside funds for innovations so that there would be resources to identify the best new advances in care.  We recognized that the children system of care and the adult system of care were started with new dedicated state funds and an evaluation of these pilot projects.  Due to their success we were able to get additional funds to expand them to other counties.

This part of the Act did not wind up being implemented until 2012.  The way it was set up each county was on its own to identify new innovative projects.  The first round of these projects are now completed.  I would have thought that we would have a detailed report on the results and a series of new models that could be recommended for replication by other counties.

That has not happened. In the meantime several new innovations have emerged which include:

  • Integrated crisis centers to divert people from both jails and hospitals
  • School county multi-tiered partnerships
  • College mental health systems
  • Workplace mental health programs
  • Prodromal phase “early psychosis” programs
  • Youth centers

This program needs to be restructured so that counties can choose from a list of innovations for which the oversight and accountability commission (MHSOAC) has provided guidelines and to include a method to have a statewide evaluation that will allow all counties to benefit and learn how to implement the program.  Allowing all counties to use innovation funds to implement the startup of any of these approved innovations should be an eligible use of funds for several years after the initial projects are completed.

As documented in our 2016 policy recommendations, the approved list should be developed through an interactive process that collects the best ideas that come to counties locally as well as those that come directly to the commission.

In addition, counties would be free to propose other projects with an understanding that they are not as likely to be approved unless they appear to be equally significant or the county has already implemented all of the programs on the approved list that are feasible for a County of its size.

Prevention and Early Intervention (PEI)

Our campaign slogan in 2004 was to go from “fail first to help first.”   This recognized that we currently have a fail-first mental health system whereby people don’t get into the system until they had hit a crisis level of failure in some other system. What we wanted was to change how those other systems worked so that they identified mental illnesses at the first possible opportunity and got people help at that stage before their condition was life-threatening or disabling. We knew very little about what would be the best strategies for this program besides the value of school-based approaches to children and early psychosis programs. Nearly all counties now have some school-based approach as well as an early psychosis program.  But there is also an enormous variety and how counties of approached this program.

I have often quoted a cardiologist who said that when he started practicing they did one hundred different things but by the time he retired there were only doing five because everyone learned what the most effective strategies were.

That is what needs to happen for the prevention and early invention program. We need to identify that small group of program elements that should be universal statewide.  I wrote a paper in 2014 which identified four of them besides early psychosis:

  • School multi-tiered approach reflecting school county partnerships with on-campus clinicians and a process to identify all at risk students as early as possible
  • Primary care and emergency room screening with immediate initiation of evaluation by a behavioral health professional
  • Workplace behavioral health programs that steer everyone to seeking help when a coworker or manager recognizes that something isn’t right
  • Internet strategies that steer people to getting help

Since writing that paper I’ve also become aware of youth centers and college programs that also should be promoted.  The MHSOAC should develop specific strategies to take each of these core PEI programs to scale, as well as identifying others that should also become universal.

As this work is done the commission should rewrite the guidelines and regulations so that all counties are incentivized to implement those programs, initially with innovations funds and then to take them to scale with PEI funds.


For prevention and early intervention, the outcome measures listed in the Act are actually the absence of bad outcomes which PEI should make happen. As it turns out, it is very hard to measure these outcomes in relationship to specific prevention and early intervention activities. A better approach is to focus on the core PEI activities and measure the extensiveness of the implementation.

  • What percentage of the people expected to develop new cases of schizophrenia each year are we capturing with early psychosis programs?
  • What percentage of people are screened in primary care?
  • Of those who screened positive what percentage get care that helps them?
  • What percentage of schools have early identification systems?
  • What percentage of workplaces have early identification systems?

These types of measurements would go a long way to demonstrating that we are making prevention and early intervention a norm instead of an exception.


The Little Hoover Commission reports in 2015 and 2016 documented the lack of strength and clarity in roles and responsibilities of the oversight commission and the Department of Health Care Services.

As we anticipate the election of a new governor in the fall of 2018, it is an appropriate time to evaluate what is working and what isn’t – and what steps will get us where we need to go, the primary issue still being a lack of data that would allow for meaningful oversight.


In the past 10 years, I have heard more complaints about this than just about any other part of the Act. The two most common problems are lack of access to all of the relevant budget details in terms of available revenues relative to past and projected expenditures, and lack of an interactive process with stakeholders after the county has developed its draft spending plan.

In addition, it is clear that the stakeholder process only works in relationship to community services and supports and that the key other interests necessary to get to the best prevention and early intervention programs are not generally part of the process.

The best PEI programs happen outside of the public mental health system. Getting major changes in these other systems requires engagement at higher levels of leadership such as County Supervisors, County Executives and Sheriffs interacting with their counterparts in schools, health plans, cities and other key partners.

Guidelines for the local stakeholder process did not differentiate between community services supports and PEI. In fact they were written in conjunction with the implementation of the community service and supports program and were never revised to reflect the changing needs for prevention and early intervention.

The guidelines clearly need to be revised to focus on the qualitative issues and financial transparency concerns regarding community services and supports.

A completely new set of guidelines needs to be written for prevention and early intervention stakeholder planning that includes ways to engage the stakeholders reflected by the composition of the MHSOAC that might include creation of local commissions with such a composition.


The primary purpose of the Act was to fully fund the AB 34/AB 2034 adult system of care program – to provide whatever it takes services to everyone who was homeless or at risk of homelessness due to a severe mental illness.  We have succeeded in making this the model of care and significantly expanding the number of people it serves.

We have failed to serve enough people to make a significant dent in the number who are homeless incarcerated or hospitalized.  Mostly this is due to the problems with 1991 realignment – both the diversion of growth to in-home support services and the lack of growth due to the recession that started in 2008. Together these problems cost mental health $700 million annually that we will never get back and now represents about $1 billion due to inflation. As a result the first $1 billion of community services and supports (just about all of the funding with $1.5 billion total revenues) simply goes to backfill this loss and does not allow for expansion.

Over time the MHSA will grow, but I don’t see that being enough to help us meet our needs unless we find a way to accomplish the second purpose of the Act which is to make prevention and early intervention the norm and not the exception.

We now know a number of key strategies for PEI that we didn’t know when we wrote the Act.  If we implement all of these strategies across the state in a comprehensive way we should be able to significantly bend the cost curve by reducing the number of people who need the higher levels of care.

It takes me back to our first focus groups in writing the Act. In each of those groups we put up on the flip charts the words “mental health”.  The facilitator asked what comes to mind when you see these words. In every group the first response was “street people”.  In the discussions that followed each group also said “but why do people have to become homeless before they get help?  Why can’t we get people help when they first show any signs of mental illness?”  We now know that we can. The question is how quickly and fully will we make that the norm.


Leave a Reply

Your email address will not be published. Required fields are marked *