Hide Notes 

Slide 1: Cost-Effectiveness of Services for Mentally Ill Homeless People: The Application of Research to Policy and Practice American Journal of Psychiatry Robert Rosenheck, MD 10/00
Slide 2: Background Information    25% of homeless Americans have a serious mental illness 500,000 Americans on average, for the past 20 years, are sleeping in shelters or on the street each and every night Complicated problems- poor physical health, past trauma, long-term poverty, isolation, lack of vocational skills, stigmatization in the criminal justice system (and even in our own minds)
Slide 3: Why is this important?    The United States is one of the most affluent nations in the world and therefore we should be able to care for our the most disadvantaged American Psychiatry has seen remarkable scientific and clinical advances to treat the mentally ill, including long acting injections and newer and safer anti-depressants The article states that we, as a nation, have failed for 20+ years to protect mentally ill citizens from drifting into homelessness
Slide 4: Inclusion and Exclusion Criteria    Excluded studies involving programs for homeless people with primary addictive disorders Only 3 experimental cost-effectiveness studies have been published, so observational outcome studies were included (potential bias) This study summarizes recent researches’ implications for helping a deeply disadvantaged and difficult to treat population
Slide 5: Applications and Strategery    The article tried to generalize these smaller selected studies to larger populations Attempted to use cost-effective analysis to help design a program that would target a “real” population of homeless people The ultimate decision is going to be made partly by the American people and partly by our willingness to place our values and attitudes in the forefront, despite knowing that treating the homeless is expensive. Not treating the homeless also has its costs.
Slide 6: Types of Interventions  Outreach, case management, and housing placement/transition to mainstream services Outreach- for people who are reluctant to seek help on their own Case management- personalized relationships that facilitate access Placement and transition- provide long-term solutions   
Slide 7: Outreach  New York Choices program- multifaceted approach:     Outreach and engagement Low-demand, day/drop in center 10 bed housing unit Community based rehabilitation Clients experienced 54% reduction in nights on the street, twice the 28% reduction among control subjects This is unlike intensive case management where program costs are usually offset by reduced inpatient treatment Problem- did not address the time and effort by outreach workers building rapport* with clients before entry into treatment or those who are “courted” but do not enter treatment (frustrating!!)   
Slide 8:     Access to Community Care and Effective Supportive Services Program 12,000 clients were contacted, only 46% ()  entered case management Only 19% of those contacted in street settings eventually entered case management () Clients, not surprisingly, improved clinically and spent less days homeless () Bottom line- more costly because for every 1 successful case, 4 were screened ($$) and engagement took 2X as long as in other studies (100 vs. 50 days) Failed encounters add to the average cost of a successfully engaged client
Slide 9: VA program, Figure 5    The government meticulously keeps records of the costs of homeless outreach (so there is a lot of data) but still they can’t seem to balance the budget VA health care costs increased from $6400 per patient per year (pppy) to $7300 pppy, not including increases of $1100pppy (for residential treatment), $315pppy (for case management) Bottom line- when effective outreach can be costly.
Slide 10: Assertive Community Treatment vs. Broker Case Management     Broker case management (just linking client to services) was the most costly at $45,000 per client over 18 months Assertive Community Treatment was the least expensive at $39,000 per client over 18 months This showed that the high costs of case management can be offset (at least partially) by reduced hospital use More importantly, I hope, assertive community treatment clients had reduced symptoms and were more satisfied with the services they received
Slide 11: Costs Incurred Over 18 Months by Clients in the Assertive Community Treatment Study, Figure 2
Slide 12: Assertive Community Treatment vs. Standard Care      Total costs were $50,000 for assertive community treatment vs. $66,000 for control subjects More importantly, there were improvements in Symptoms, Life Satisfaction, and Health Status Reduction in costs ($16,000 per patient) was due almost entirely due to reduction in inpatient costs Conclusion- After the high cost outreach phase, assertive community treatment is better because it is more effective and has fewer inpatient care and costs. Problem- little external validity as these studies included patients from Baltimore and St. Louis who had especially high inpatient costs. You can show more savings for high-cost clients than for low-cost clients.
Slide 13: Health Care and Societal Costs Incurred over 12 Months with Assertive Community Treatment, Figure 3
Slide 14: A More Representative Sample    Previous studies were compared and adjusted for inflation At the 90th percentile in terms of cost, the VA patients and Access clients had inpatient costs of $32,000 and $25,000 pppy, which approached the cost of the inpatient stays for the Baltimore and St. Louis groups Conclusion- it is harder to save funds with a more representative sample since inpatient costs only represent a small percentage of overall costs.
Slide 15: Boston Evolving Consumer Households Housing Project    Studied the 3rd type of service generally provided: Placement and Transition of the Mentally Ill and is focused on long-term solutions The philosophy was that groups of homeless Americans would live together initially with substantial support from staff and then develop their own mutually supportive communities that would survive on their own without staff That brings me to my next topic…
Slide 16: The Communist Manifesto by Karl Marx and Frederick Engels
Slide 17: Boston Study      118 randomly assigned to an apartment with an off-site case management support OR an evolving consumer household (community residence with a house manager) Housing results: 92% of all nights in the community residence, 83% for independent living (statistically significant) Improvement in health status and life skills 8% for community residence and 2% for independent apartment with off-site case manager (statistically significant) Costs were $56,000 for the community residence and $30,000 for the independent living arangement Inpatient costs were less in the community residence ($9,000/yr) compared to the independent living group ($12,500/yr), but this was not enough to overcome the larger cost of assisted living  Patients who had their own apartment cost less overall (good) but had more inpatient costs (bad economically and possibly symptomatically)
Slide 18: Costs Incurred over 12 months in the Boston Evolving Consumer Households Project, Figure 5
Slide 19: Transition to Community Housing   Specialized homeless service that transitioned the mentally ill homeless into housing and mental health services “Critical Time Intervention” consists of 3 phases:    a Accommodation phase (month 1-3) where there are numerous home visits and link clients with providers a Tryout phase (months 4-7) where clients use services a Termination phase (months 8-9)
Slide 20: Data from Critical Time Intervention    Over 18 months, critical time intervention clients spent an average of 30 night sleeping on the street compared to 90 for controls- 1 out of 20 nights versus 1 out of 6 nights and exhibited greater improvement in negative psychiatric symptoms Critical time intervention clients had slightly more hospital days (42 versus 38), ER visits (1.5 versus 1.2), more outpatient visits (30 versus 17) and more day program visits (74 versus 69) Conclusion- increased costs, but not much more in terms of hospital stays**. When we discharge a pt, where do they go?
Slide 21: Conclusions   500,000 Americans on average, for the past 20 years, are sleeping in shelters or on the street each and every night New York Choices- Clients experienced 54% reduction in nights on the street, twice the 28% reduction among control subjects.     Assertive Community Treatment- the high costs of case management can be offset (at least partially) by reduced hospital use but more importantly assertive community treatment clients had reduced symptoms and were more satisfied with the services they received Critical Time Intervention had only 4 more hospital stays over the course of 18 months. Patients do go somewhere. Independent Apartment living with an separate case manager is less costly than community residences (good or bad?) The standard for determining the value of a day not sleeping on the street or actively hallucinating is determined by the public’s willingness to pay for that day. One study found that people in the US were willing to pay 301$ in 199 inflation adjusted dollars.
Slide 22: Implementations      If you target “frequent flyers”, you can save the most $$ as there is more to be saved for the “high-cost” patient. One study showed that increased public support was associated with superior housing outcomes (but they had negotiated with the SSA) and jumped to the head of the line in terms of housing. So just income support doesn’t seem to work, there must be a combination of income and housing support. Most policy analysts state that increasing housing and income support for the mentally ill homeless is the answer, but the use of services and their effectiveness is understudied. Innovative programs are modestly more effective and may be much more expensive. The only clientele that you could say “invest money into there programs, as there will be better outcomes, and you will save $ that you can use to treat others” are the most resource intensive 10% of clients mentioned, the high-cost patient.
Slide 23: Famous Quotes about CostEffectiveness and Economics  "A surplus means there'll be money left over. Otherwise, it wouldn't be called a surplus." -Unknown President