Impact of homelessness on individuals with schizophrenia

Individuals with schizophrenia experience higher rates of co-occurring medical disorders, substance use disorders, other psychiatric disorders, premature mortality,Reference Olfson, Gerhard, Huang, Crystal and Stroup ^{12 ,} Reference Tsai and Rosenheck ¹⁴ and suicide when compared to the general population.Reference Palmer, Pankratz and Bostwick ¹⁵ For the unhoused person with schizophrenia, these adverse consequences are heightened as they have limited access to medical and mental health treatment. It is no surprise, therefore, that unhoused individuals with schizophrenia face more frequent psychiatric hospital readmission rates, an emotional burden for the individual as well as a financial burden for society. In their retrospective study of 207 patients who had been psychiatrically hospitalized, Lorine et al.Reference Lorine, Goenjian, Kim, Steinberg, Schmidt and Goenjian ¹⁶ evaluated 207 patients who were discharged and followed up at 3 different time periods to examine readmission rates. The time frames were readmission within 15 days (Group 1), readmission within 3–6 months (Group 2), and not being readmitted from at least 12 months (Group 3). Of this sample, 50% had schizophrenia or schizoaffective disorder and 24% of the sample were homeless. The study found that having a diagnosis of schizophrenia or schizoaffective disorder increased the odds of being readmitted within 15 days versus not being readmitted within 12 months by nearly 18 times. For those with a diagnosis of schizophrenia or schizoaffective disorder, being homeless increased their odds of readmission within 15 days versus not being readmitted by nearly 30 times. These results highlight the particular risk that having both a schizophrenia spectrum disorder and being unhoused plays in hospital readmission.

Unhoused individuals with schizophrenia are also at an increased risk of becoming both perpetrators and victims of violence. A diagnosis of schizophrenia in and of itself confers an increased risk of violence toward others when compared to the general population.Reference Wallace, Mullen and Burgess ¹⁷ Untreated schizophrenia is characterized by active psychosis, which typically includes symptoms of paranoia, hallucinations, and false beliefs about others and one’s environment. In an analysis of 204 studies examining the relationship between psychopathology and aggression, Douglas et al.Reference Douglas, Guy and Hart ¹⁸ found that psychosis was the most important predictor of violent behavior in an individual. Although significant research indicates that unhoused individuals without schizophrenia have higher rates of violence and criminal offending,Reference Gonzales, Jetelina and Roberts ^{19 ,} Reference Snow and Anderson ²⁰ does the combination of both circumstances heighten the risk of criminal offending further? The answer is yes. Research by Nilsson et al.Reference Nilsson, Laursen, Andersen, Nordentoft and Fazel ²¹ indicates that individuals with a severe mental illness (eg, schizophrenia or bipolar disorder) who are unhoused have higher rates of violence than those who are housed.

Unhoused individuals with schizophrenia are at increased risk of being victims of violence in addition to their increased risk of victimizing others. In their study, Roy et al.Reference Roy, Crocker, Nicholls, Latimer and Isaak ²² reviewed 21 studies investigation the relationships of persons experiencing homelessness with serious mental illness (PEHSMI) to violence and victimization. The authors reviewed 15 studies specific to contacts with the criminal justice system and 6 studies specific to the prevalence of victimization. Their analysis indicated that PEHSMI had lifetime arrest rates ranging from 63% to 90%, lifetime conviction rates ranging from 28% to 80%, lifetime incarceration rates ranging from 48% to 67%, and lifetime victimization rates ranging from 74% to 87%.

Interventions for unhoused individuals with schizophrenia

With the goal of implementing the most effective treatment approaches for unhoused individuals with schizophrenia or other psychotic disorders, 2 competing models have been implemented: Treatment First versus Housing First (HF). The Treatment First model, also known as the “Continuum of Care” model, was the dominant model for managing psychiatric patients who had been discharged into the community during the deinstitutionalization movement of the 1960s and 1970s. This model typically consists of a linear model of care, where individuals with a psychiatric disorder have a step wise progression of services that help prepare the individual for independent living, a process known as “housing readiness.”Reference Tsemberis, Moran, Shinn, Asmussen and Shern ²³

Under the Treatment First model, the unhoused individual is typically expected to demonstrate sobriety and readiness to accept mental health treatment while in temporary housing. In the Treatment First model, case managers evaluate whether the person has sufficient life skills to live without onsite supervision.Reference Dordick ²⁴ Once the individual is deemed “housing ready,” then independent housing is considered. Individuals who refuse to engage in psychiatric treatment, who continue to use substances, have a history of violence or incarceration, or problematic behaviors are not typically placed in independent housing under this model.

In contrast, HF model emphasizes the importance of first being placed in permanent housing regardless of whether the person is actively psychotic, using alcohol or drugs, agrees to occupational rehabilitation, or accepts recommended treatment interventions. The HF model’s origins can be traced to the Consumer Preference Supported Housing (CPSH) model that arose from a private nonprofit social services organization in New York City known as Pathways to Housing, Inc. Under this model, housing is considered a right for all individuals and interventions are client centered. The CPSH model is the precursor for the HF model, and its foundation includes the following tenantsReference Tsemberis and Asmussen ²⁵:

1. 1. Unhoused individuals with a SMI can successfully live independently in housing of their choice with appropriate supports;

2. 2. The individual chooses their own housing;

3. 3. Housing is rented from a community landlord who does not provide the support services;

4. 4. The individual does not lose their housing when they are in a clinical crisis, such as a substance use relapse or psychotic episode;

5. 5. Services are provided by a community ACT team and available 24 hours a day;

6. 6. The individual selects the type, frequency, and order of services chosen;

7. 7. No form of treatment is required, including sobriety or medication compliance. Instead, the harm reduction model is used to address alcohol and drug abuse.

In their 2004 research examining the longitudinal effects over a 2-year period of a HF program for homeless individuals with severe mental illness, Tsemberis et al.Reference Tsemberis, Gulcur and Nakae ²⁶ compared 126 participants assigned to the Continuum of Care model to 99 participants assigned HF model program. Fifty-three percent of those enrolled in the study had a psychotic disorder diagnosis. Although those placed into the HF program were able to maintain their independent housing over the 24-month period, no differences were found in substance use or psychiatric symptoms between HF versus the Continuum of Care participants.

In their systematic review of 72 articles, Aubry et al.Reference Aubry, Bloch and Brcic ²⁷ examined the effectiveness of permanent supportive housing and income assistance for homeless individuals in high-income countries. According to these researchers, most of the studies on permanent supportive housing they reviewed included individuals with severe mental illness who had been unhoused. The authors concluded that both permanent supportive housing and income assistance significantly improved housing stability at 6 years of follow-up. However, this systematic review did not provide evidence that unhoused individuals with permanent supportive housing had improvements in their mental health or substance use compared with controls. In other words, HF may increase housing stability but has not been shown to decrease symptoms of SMI. This finding should inform public policy that to decrease the psychiatric disease burden, antipsychotic medications should be combined with HF to improve the mental health outcomes of unhoused individuals with schizophrenia and other forms of active psychosis.

In their subsequent study, Loubiere et al. Reference Loubiere, Lemoine and Boucekine ²⁸ studied the effects of the HF model among homeless people regarding housing stability, quality of life, healthcare use, mental symptoms, and addiction issues. Researchers examined data from a randomized controlled trial involving homeless or precariously housed adults with severe mental illness from 4 French cities. A total of 703 participants were selected for this study, and all had a diagnosis of either bipolar disorder (30%) or schizophrenia (70%) according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision (DSM-IV-TR). In addition to having a bipolar disorder or schizophrenia disorder diagnosis, participants had to have at least 1 of the following: 1. two or more psychiatric hospitalizations in the past 5 years; 2. co-occurring alcohol or substance use disorder; or 3. history of arrest or incarceration within the previous 2 years. Participants who could not provide informed consent for the study were excluded.

The study participants were divided into 2 groups for comparison outcomes: a treatment-as-usual (TAU) group (n = 350) and a HF group (n = 353). For those who were assigned to the TAU group, their housing situations varied, with some individuals living on the streets, with friends or families, or in slums. TAU interventions involved preexisting programs and services for the homeless. These services included outreach teams, day-care facilities, access to emergency shelters and transitional shelters, residential facilities with medical accommodations if needed, and independent housing. In contrast, the HF model involved the provision of independent housing with housing subsidies and assertive community treatment provided by a mobile support team. Both groups were followed over a 48-month period. Both the TAU and HF groups improved in measures of recovery although no statistical difference in recovery outcomes was noted. The HF group compared to the TAU group had a lower use of hospital services. However, no significant differences were found between the 2 groups related to self-reported mental symptoms or substance dependence. In fact, HF participants experienced higher alcohol consumption between baseline and 40 months. The findings in this study that HF did not reduce mental health symptoms replicate the results from the Aubry et al.’sReference Aubry, Bloch and Brcic ²⁷ meta-analysis described above.

Whether or not a homeless individual with schizophrenia is housed, does compulsory treatment, most often involuntary medication administration, improve outcome? Compulsory community treatment orders (CTOs) are legally mandated orders that require psychiatric treatment for identified individuals with severe mental illness who do not voluntarily accept treatment and are considered a risk of harm to self or others or are unable to care for themselves. Failure to follow requirements in CTOs may result in involuntary psychiatric hospitalization. In some jurisdictions, CTOs require a showing that the individual lacks medical decision-making capacity regarding the use of psychotropic medications.Reference Beaglehole and Tennant ²⁹

In the 2017 Cochrane review of compulsory community treatment (CCT) for people with mental illness, Kisely and CampbellReference SR and Campbell ³⁰ reviewed 3 randomized controlled clinical trials of CCT compared with standard care for people with SMI. The authors concluded that their review did not demonstrate that those receiving CCT had differences in improvement in the areas of service use, social functioning, or quality of life compared with those who received voluntary care or brief supervised discharge. However, those receiving CCT were less likely to be victims of violent or nonviolent crime.

CTOs are considered controversial with debated pros and cons of their usage. Suggested benefits of CTOS include earlier treatment intervention that helps prevent mental health deterioration, increased involvement of family and monitoring clinicians, decreased recurrent hospitalizations, decreased interaction with the criminal justice system, and decreased victimization. Concerns regarding the use of CTOs include therapeutic alliance disruption, adverse medication side effects, stigmatization, disproportionate application to people of color or indigenous populations, and a resulting reluctance of the patient to seek future treatment.Reference De Waardt, van Melle and GAM ³¹

Perhaps CTOs are more effective for some individuals than others, depending on the diagnosis. For example, Beaglehole et al.Reference Beaglehole, Newton-Howes, Porter and Frampton ³² reviewed nearly 15 000 patients in New Zealand who were placed on a CTO over a 10-year period between January 2009 and December 2018. This study examined the number of psychiatric inpatient admissions per year for individuals on CTOs for a range of psychiatric disorders. These researchers found that the use of CTOs for individuals with a psychotic disorder resulted in reduced hospitalization admissions and frequency. In contrast, individuals with dementia, bipolar disorder, major depressive disorder, and personality disorder had more frequent hospital admissions that were also of longer duration. The authors concluded that compulsory treatment for individuals with psychotic disorders appeared effective in reducing psychiatric hospital admissions and, therefore, relevant in decreasing the disease burden of psychosis.

Optimizing interventions for unhoused individuals with schizophrenia

Unhoused individuals with schizophrenia face an array of challenges, not the least of which is survival. With increased mortality and suicide rates, what are common sense approaches to caring for these persons who may not want and even refuse treatment? First, practitioners and policymakers should be familiar that many persons with schizophrenia suffer from anosognosia, thereby limiting their ability to appreciate their symptoms and robbing their rational capacity to accept or refuse treatment. Research indicates that between 30 and 50% of patients with schizophrenia lack insight as a prevalent feature of their disorder.Reference David ³³

When assessing for decisional capacity to accept or refuse treatment, evaluating the individual’s insight into their mental illness may assist in determining whether treatment refusal is linked to an unawareness that they are psychotic and could benefit from treatment. A consensus definition for assessing insight includes addressing the following questionsReference Thompson, McGarry and Harrigan ³⁴:

1. 1. Is the person aware that they have a mental illness?

2. 2. Does the person understand the need for treatment?

3. 3. Is the person aware of the potential adverse social consequence related to their mental disorder?

4. 4. Is the person aware that they have symptoms?

5. 5. Is the individual able to attribute their symptoms to a mental disorder?

One widely used tool to assess insight in clinical trials and epidemiological studies is the Scale to Assess Unawareness in Mental Disorder (SMUD).Reference Amador ³⁵ An abbreviated and more practical version of the SUMD has been developed to assist clinical evaluators assess a patient’s insight into their mental illness. In contrast to the full SMUD, which has 74 items, the abbreviated SMUD has only 9 items which are rated on a severity scale. This abbreviated version rates the following 9 items as related to their mental awareness:

1. 1. mental disorder;

2. 2. consequences of a mental disorder;

3. 3. effects of drugs;

4. 4. hallucinatory experiences;

5. 5. delusional ideas;

6. 6. disorganized thoughts;

7. 7. blunted affect;

8. 8. anhedonia; and

9. 9. lack of sociability.

This abbreviated version has demonstrated that it is a valid instrument for measuring insight in patients with schizophrenia and can accurately assess insight in clinical settings.Reference Michel, Baumstarck and Auquier ³⁶

Second, when balancing liberty interests in refusing effective medication treatments for schizophrenia versus the treatment benefits, the reality that medications decrease psychotic symptoms and improve outcomes should not be ignored. Large meta-analyses have demonstrated that oral antipsychotics effectively decrease acute psychotic symptoms.Reference Huhn, Nikolalopoulou and Schneider-Thoma ³⁷ Research also indicates that long-acting antipsychotic injectables are efficacious in treating acutely psychotic individuals with schizophrenia-spectrum disorders.Reference Vita, Pollini and Canozzi ³⁸ Because many individuals with schizophrenia have poor medication adherence, the use of a long-acting injectable helps optimize the delivery of an effective medication for often debilitating symptoms. Failure to provide antipsychotic treatment as early as possible has been associated with negative outcomes. In their research studying 2 longitudinal cohorts of patients with first-episode psychosis, Drake et al.Reference Drake, Husai and Marshall ³⁹ found that a long duration of untreated psychosis (DUP) was associated with a reduced treatment response over time. DUP represents the time between onset of the first threshold psychotic episode and the initiation of treatment. Longer DUP has demonstrated numerous negative outcomes including less likelihood of symptoms going into remission along with a decreased quality of life and level of functioning.Reference Frischherz, Conus and Golay ⁴⁰ When faced with individuals with schizophrenia who refuse treatments that assist in bringing them closer to reality, psychiatrists should be knowledgeable about the negative impact of delaying treatment so that they can meaningfully inform the relevant decision-maker responsible for treatment refusal overrides. As highlighted above, individuals with schizophrenia are at a much greater risk of early death, completed suicides, and becoming unhoused. These adverse consequences of remaining untreated are relevant when considering compulsory medication to maximize positive treatment outcome.

Third, although the HF approach has shown that individuals with a severe mental disorder who are placed into housing have less days homeless, the research has not demonstrated that those with SMI experience actual symptom reduction with HF alone. Although being placed into housing may decrease the stress and trauma of living on the streets, the HF model by itself is not an effective treatment for the psychotic symptoms of schizophrenia. Combining the HF approach with consideration of compulsory antipsychotic treatment in symptomatic treatment refusers is essential to prevent deterioration and maximize functioning. As noted above, compulsory treatment orders for individuals suffering from psychosis have demonstrated their utility in decreasing hospitalizations for individuals experiencing psychosis.

Fourth, intensive case management and assertive community treatment provide meaningful support and assistance for individuals with schizophrenia. However, for some individuals with schizophrenia, this level of care does not adequately manage their needs or symptoms. Lamb and WeinbergerReference Lamb and Weinberger ¹³ emphasize the need for more 24-hour structured care facilities as part of the community mental health system. This level of community care will assist those individuals whose psychosis is refractory to treatment as well as decrease the risk of diversion into the criminal justice system. Future policymakers and stakeholders should recognize that current long-term care facilities are not the equivalent of the “snake pit” hospitals of the past, whose horrific conditions understandably played a role in deinstitutionalization of the mentally ill. Moving forward, such 24-hour treatment facilities will likely have an important role in the continuum of care for individuals with schizophrenia who are unhoused or at risk for becoming homelessness due to their disability.

Summary

Individuals with schizophrenia are at an increased risk of becoming chronically homeless. While attempting to survive on the streets or in a shelter, they face an onslaught of challenges, from being personally victimized, diverted into the criminal justice system, developing serious untreated medical complications, and dying prematurely. Being untreated and actively psychotic represents its own form of psychic torture. What “works best” includes the following key principles:

1. 1. A rational balance of the individual with schizophrenia’s liberty interests with restoration of their sanity and dignity through implementation of compulsory treatment orders when indicated.

2. 2. A recognition that the HF approach alone for individuals with active symptoms of schizophrenia does not address the negative long-term medical and societal outcomes of untreated psychosis. More structured interventions with required treatment involvement are necessary as part of the permanent housing approach.

3. 3. The Harm Reduction approach for individuals with schizophrenia and a co-occurring substance use disorder is unlikely to be effective and may actually perpetuate psychosis, particularly with substances such as cannabis, stimulants, and hallucinogens.

Psychiatrists and mental health professionals play a crucial role in educating policymakers, judges, and other stakeholders that effective treatment and medication therapy work. Society does not have to ignore psychotic humans in despair on the street under the guise of respecting them. In the novel “My Several Worlds,” Pearl S. Buck may have said it best when she writes “The test of a civilization is in the way that it cares for its helpless members.”Reference Buck ⁴¹ Moving forward, we should better care for members of our society who need our help. We should pass this basic test of civilization.