Article Text

Download PDFPDF

A “critical time” intervention reduced homelessness in inner city men with psychiatric problems
  1. Philip Timms, MRCPsych
  1. Guy's and St. Thomas's UMDS London, UK

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

OpenUrlCrossRefPubMedWeb of Science


To evaluate a “critical time” intervention for the prevention of recurrent homelessness in inner city men.


Randomised controlled trial with 18 month follow up.


Community based study in New York City, USA.


96 men (60% ≥35 y of age, 74% African American) who were discharged to community housing from an on site psychiatry programme in a New York City men's shelter. All patients had major mental illness, such as schizophrenia or bipolar psychoses. Housing ranged from intensively supervised community residences to single room occupancy hotels with on site social services. Follow up was 98%.


Patients being resettled were allocated to a critical time intervention (CTI) (n=48) or to usual services only (USO) (n=48). The CTI involved the patient having the support of a CTI worker who was experienced in working with homeless people and who facilitated the transfer of care from the shelter to other caregivers in the community. Particular areas of focus were medication adherence and money management. The CTI was implemented for 9 months followed up by 9 months of USO.

Main outcome measure

Number of homeless nights estimated using monthly face to face interviews.

Main results

The mean number of homeless nights was 30 in the CTI group and 91 in the USO group (95% CI for the 61 night difference 19 to 105, p=0.003). Homelessness lasting >54 nights occurred in 21% of patients receiving the CTI compared with 40% of patients receiving USO (p=0.045) (table).

Critical time intervention (CTI) v usual services only (USO)*


A critical time intervention that fostered long term support in the community reduced subsequent homelessness in men with psychiatric problems who were discharged from a shelter.


The issue of resettlement failure is important. Even using well resourced specialist teams, the resettlement placements of one fifth of mentally ill homeless people will have failed after 1 year.1 This is wasteful and demoralising for patients and workers. The homeless milieu has traditionally been hostile to tightly constructed interventions, but this study by Susser et al provides a refreshing attempt to evaluate a well defined social support intervention with a clearly defined outcome measure over a substantial period of time. The follow up rate was impressive and shows what can be achieved with a committed team of workers who are familiar with the homeless environment.

Many mental healthcare workers working with homeless people combine several roles in a rather confusing manner—crisis intervention, acute assessment, case management, social support, and rehabilitation. Support work has traditionally been seen as a continuing activity that does not require specialised skills. This has led to a lack of clarity about its potential effect and subsequent neglect. In the team for homeless people with which I work, our 3 support workers were able and acquired assessment skills. However, because we could not quantify what they achieved in their support role, they were reallocated to assessment work to the detriment of our service. I therefore welcome this study because it shows the potentially powerful effect of a particular kind of support work, involving personal support and what might be called “community reconnection”. This was active, focused, and effective in a measurable way, with effects lasting beyond the period for which it was provided. Furthermore, the advantage of using specific personnel is that without conflicting demands on their time it is more feasible to adjust their level of input according to individuals' needs. It is also a way of more clearly estimating the resources that are needed to achieve successful resettlement.



  • Source of funding: National Institute of Mental Health.

  • For article reprint: Dr E Susser, Columbia University/New York State Psychiatric Institute, 722 West 168th Street, Box 24, New York, NY 10032, USA. Fax +1 212 795 9768.

  • Adapted from an abstract published in Evidence-Based Medicine 1977 Sep-Oct;2:148.