Mental Health National Service Framework (Wales) edition

Health Evidence Bulletins - Wales

Literature search: Comprehensive systematic search to November 2003 plus selected update searches to December 2004 as advised by review groups 

Chapter 5. Commissioning Services

Effective services must be jointly planned, commissioned and delivered in an efficient co-ordinated manner in order to provide responsive, seamless care.

National Service Framework: key action 17

Adult mental health services: a national service framework for Wales. Welsh Assembly Government 2002

Effective services must be planned, designed, and delivered to meet the needs of the population. They should take an epidemiological approach following formal comprehensive needs assessment at both national and local levels and a gap analysis of service provision. [Key action 17 paragraph 20.2]

What evidence is available on conducting needs assessment for service planning?
What are the needs of different population groups?

The Statements The Evidence
5.1 Needs assessment

5.1a The routine use of outcomes measures and needs assessment tools is, as yet, unsupported by high quality evidence of clinical and cost effectiveness. Clinicians, patients and policy makers alike may wish to see randomised evidence before this strategy is routinely adopted. No randomised data were found which addressed the specified objectives. one unpublished and one ongoing trial was identified.i

i. Gilbody SM, House AO, Sheldon TA. Outcome measures and needs assessment tools for schizophrenia and related disorders. The Cochrane Database of Systematic Reviews 2003, Issue 1.
clsysrev/articles/CD003081/frame.html [accessed 29/07/05
(Type I evidence - systematic review to establish the value of the routine administration of outcome measures and needs assessment tools and the feedback they provide in improving the management and outcome of patients with schizophrenia and related disorders. Literature search to 2002.)

5.1b Routine HRQoL measurement is a costly exercise and there is no robust evidence to suggest that it is of benefit in improving psychosocial outcomes of patients managed in non-psychiatric settings. Major policy initiatives to increase the routine collection and use of outcome measures in psychiatric settings are unevaluated. All the instruments used included an assessment of mental well-being, with specific questions relating to depression and anxiety. The routine feedback of these instruments had little impact on the recognition of mental disorders or on longer term psychosocial functioning. While clinicians welcomed the information these instruments imparted, their results were rarely incorporated into routine clinical decision making. No studies were found that examined the value of routine assessment and feedback of HRQoL or patient needs in specialist psychiatric care settings or examined the costs and resource use associated with routine outcome measurement.i

i. Gilbody SM, House AO, Sheldon T. Routine administration of Health Related Quality of Life (HRQoL) and needs assessment instruments to improve psychological outcome - a systematic review. Psychological Medicine 2002; 32(8): 1345-56
(Type I evidence – narrative systematic review to examine the addition of routinely administered measures of Health Related Quality of Life (HRQoL) to improve psychological care and outcome of people being managed in non-psychiatric care settings (n=9) and those in specialist psychiatric care settings (no studies were found). Literature search to 2000.)


5.1c Systematic assessment of needs with research instruments to contribute to locality service planning is possible without a major research grant, but involves compromises on established research designs. In the Gloucester locality, patients with functional psychosis were largely in contact with mental health services, were in employment, were disproportionately looked after by a few city centre general practitioners, and high levels were in supported accomodation. 474 cases were identified, including 403 in contact with mental health services. Staff Camberwell Assessment of Need Short Appraisal Schedule (CANSAS) data were collected for 225 patients, with a mean rating of 7.0 met and 3.6 unmet needs per patient.i

A comparison between needs ratings assessed by staff and by the patients found that the levels of agreement between staff and patient ratings of need were significantly higher than previously reported. Possible reasons for the higher concordance were the use of keyworkers who knew patients well. Further, keyworkers may have been influenced in their assessment of need by awareness of patients’ perceived need. Joint needs assessment may strengthen the therapeutic alliance, improve our understanding of priority needs and aid in service development. Work is needed to ensure that care is targeted explicitly towards unmet need.

Unmet need was rated most highly in social/relationship domains by patients and keyworkers. Levels of staff agreement between patient and staff ratings were ‘substantial’ in all domains except safety to others, where agreement was ‘fair’ and in which staff appeared to perceive higher risk than patients.ii

Caveat: The response rate to the study was low, of the 403 patients with psychosis in contact with mental health services, CANSAS ratings were returned by keyworkers for 225 patients (55.8%).

i. Macpherson R, Haynes R, Summerfield L, Foy C, Slade M. From research to practice. A local mental health services needs assessment. Social Psychiatry  & Psychiatric Epidemiology 2003; 38: 276-281
(Type IV evidence – study to investigate the level of need among those in contact with mental health services in Gloucester City, and to identify the compromises which are involved in routine collection of data to inform services, compared with specifically funded research studies. The needs of those 225 patients with functional psychosis in contact with mental health services were assessed by keyworkers using the Camberwell Assesment of Need Short Appraisal form (CANSAS).)
ii. Macpherson R, Varah M, Summerfield L, Foy C, Slade M. Staff and patient assessments of need in an epidemiologically representative sample of patients with psychosis. Social Psychiatry and Psychiatric Epidemiology 2003; 38: 662-667

(Type IV evidence –  study to assess the relationship between patient and staff perception of need and the association between unmet need and study variables when assessed by patients and staff, respectively. 225 cases of functional psychosis in contact with a specialist mental health health team, were assessed using CANSAS by keyworkers and also by the patients.)


5.1d The 5 continuity-of-care measures are relatively easy and inexpensive to generate using administrative data. The 5 continuity-of-care measures may be useful for identifying individuals at risk for poor outcomes and for evaluating the ability of public service systems to keep clients engaged in care over time. All 5 continuity measures demonstrated good construct validity by the fact that homelessness was significantly (p < 0.001) and substantially associated with lower continuity of care.i

Caveat: Participants were compensated $25 for interview participation. As the study was conducted in America with a predominantly African American sample (59%) the results may have limited generalisbility to the UK .

i. Fortney J, Sullivan G, Williams K, Jackson C, Morton SC, Koegel P. Measuring continuity of care for clients of public mental health systems. Health Services Research 2003; 38(4): 1157-75
(Type IV evidence – retrospective cross-sectional study of 261 participants living in Houston (59 homeless, 83% male, mean age 41-years; 202 housed, 49% male, mean age 43 years) who completed face-to-face structured interviews to measure sociodemographic and clinical data. 5 distinct conceptual dimensions of continuity of care were defined: timeliness, intensity, comprehensiveness, stability, and coordination.)


5.1e Guidance is available providing information on commissioning adult mental health services. The commissioning process is referred to as the process of determining whether services are needed and how they are provided; whether services need to be changed or new services developed. The information provided includes; the commissioning process, partnership and integrated care arrangements, service user and participation, joint commissioning, and planning structures and service integration.i i. Mental Health Policy Implementation Guidance. Commissioning Adult Mental Health Services. Cardiff: Welsh Assembly Government, 2002. commissioning-adult-mh.pdf  [accessed 29/07/05] (Type V evidence – expert opinion.)
5.2 Needs of the population, and barriers to accessing mental health services
Socially deprived

5.2a Mental health differences in Wales are partly explained by the level of regional social deprivation. Of the total variance in the mental health index, 1.47% occurred at the regional level (95%CI 0.56 – 2.38). Adjustment for individual characteristics did not explain the between-region variation. A higher area deprivation score waas associated with a higher score on the mental health index.i

Caveat: Mental disorder was assessed using a five-item self-reported measure. The cross-sectional design does not examine issues of causality or duration of exposure.

i. Skapinakis P, Lewis G, Ricardo A, Jones K, Williams G. Mental health inequalities in Wales, UK: mulit-level investigation of the effect of area deprivation. British Journal of Psychiatry 2005; 186: 417-422 [accessed 29/07/05]
(Type IV evidence – analysis of data on 26,710 individuals from the 1998 Welsh Health Survey, which collected information on physical and mental health of adults aged 18 years and over. This analysis investigates psychiatric morbidity using the mental health index included in one of the surveys instruments (the Short Form 36 health survey) and deprivation estimated with the Welsh Index of Multiple Deprivation.)

5.2b Access to care for mental illness or substance abuse from a specialist is especially limited in poor communities, both for poor and non-poor individuals. Persons in poorer communities do not appear less likely to receive outpatient care or behavioural health care from a primary care physician, but they are less likely to receive speciality health care. In high income communities more than half of individuals who receive behavioural health care receive specialty behavioral healthcare, compared to just over one-third of individuals in low income communities. The 16 percentage point difference represents a 45% difference in the proportion of users accessing specialty behavioural health care in wealthy communities compared to poor ones.i

Caveat: This American study may have limited generalisability to UK settings.

i. Gresenz CR, Stockdale SE, Wells KB. Community effects on access to behavioral health care. Health Services Research 2000; 35(1pt2): 293-306
(Type IV evidence – modelling study in the US, using cross-sectional household survey data. Data were modelled using a random intercept model to estimate the influences of community-level factors on access to any outpatient care, any behavioural health care conditional on having received outpatient care and any specialty behavioural health care conditional on having received behavioural health care.)


Homeless people
See also Section 3.2 and 5.2

5.2c It can be difficult to determine the exact relationship between health and homelessness because some problems, particularly mental health problems, may sometimes pre date or even lead to homelessness as well as being caused or exacerbated by experiencing homelessness. Two-thirds of those interviewed had been homeless for up to 6-months (40% up to 3 months). Nearly 30% of those surveyed said they had concerns about or problems with mental health. The nature of those problems mainly included treatment for depression, panic attacks and suicidal tendencies and mood swings. 37% of respondents reported that they were in contact with health services on an ongoing bases. Contact appears primarily with a doctor at a GP practice, a doctor at hospital and with a psychiatrist.

Mental health issues were raised by a significant number of service providers as a key area of support needed by those homeless people with whom they come into contact. Professional concern about the “potential scale of mental health problems faced by homeless people” is high. Access to services often appears to be somewhat dependent on a “network of referrals”. The need for referrals to consultant psychiatric services, through a GP and setting up appointments can be a barrier. A high proportion of those service providers interviewed felt that some health needs of people without a permanent address are unmet. Users often do not wish to be ‘seen’ at GPs or specialist providers (especially within their own communities) and some GPs are not willing to let those people without a permanent address register.i

Caveat: Methods of data collection and analysis have not been reported for either qualitative or quantative data.

i. Adcook G. Health and homelessness in West Lothian. Newcastle-upon-Tyne: Market Research UK Limited, 2003
(Type IV evidence – qualitative and quantative survey in West Lothian
, UK. Face-to-face interviews were conducted with 97 homeless persons (48% ages 16-24 years; 40% aged 25-44 years; 3% aged >60 years; 55% male) and 28 semi-structured interviews were completed by key service providers (statutory bodies, service provider agencies and volunatry organisations).)


5.2d Homeless patients were more symptomatic and behaviourally disturbed than controls. They were significantly more likely to have a criminal history (5 cases reported being arrested in the 6-months prior to interview compared with no controls) and to be identified by key workers as having problems related to substance use (13 cases reported alcohol-related problems in the 6 months prior to interview compared to 3 controls; 14 cases reported non-prescription drug-related problems compared to no controls). Homeless patients were less likely to have been born in Birmingham (19 cases versus 32 controls, p=0.004) and to have ongoing contact with childhood carers (n=19 versus n=31 controls, p=0.004) but despite being less aware of the need for treatment, uptake of psychiatric care was comparable with that of controls.i

i. Commander MJ, Odell SM. A comparison of the needs of homeless and never homeless patients with psychotic disorders. Journal of Mental Health 2001; 10(4): 449-56
(Type IV evidence – matched case control study of homeless and never homeless patients with psychotic disorders in Birmingham. 39 matched pairs (mean age 38, 92.4% male) with a clinical diagnoses of schizophrenia or other psychotic disorder were interviewed. Clinical characteristics, insight and service use were measured as primary outcomes.)


5.2e The young homeless population is primarily male, by a ratio of around 2:1. Minority ethnic groups are under-represented in the ‘roofless’ population, but over-represented in the overall homeless population. 30% single homeless people have been in care and 20% of care leavers experience some form of homelessness within 2 years of leaving care. Young people with low confidence and self-esteem are reluctant to make full use of public services. Homeless young people on the one hand, and professional and lay people on the other, may harbour misconceptions about the other party. Housing options remain narrow for young people; for young people affected by mental health problems they are even narrower and when accommodation is made available, it may be at the bottom end of the housing market. Services must develop the ability to serve young people with multiple needs; co-morbidity is a strong feature of homeless young people with mental health difficulties. Services still fail to reach young people with the most pressing problems; more assertive outreach work is needed.i

Caveat: The methods used to conduct qualitative interviews, and collect and analyse data have not been fully reported.

i. Stephens J. The mental health needs of homeless young people Bright futures: working with vulnerable young people. London: Mental Health Foundation, 2002
(Type IV evidence – literature review and qualitative study to examine the mental health needs of young (11-25 years) homeless people. 2 focus groups were held with young people involved with Barnardo’s services and with direct personal experience of homelessness or insecure accommodation (n=16). 1 additional individual interview was conducted, on the young person’s request. Literature search date from 1990 onwards. Last search date unknown)


5.2f Although future research is needed to inform the systematic planning and implementation of services for homeless people with mental health problems, health and social services planners and providers should, in the meantime, be prepared to take risks in order to devise an innovative, effective, and nonstigmatising service response for this doubly disadvantaged group of people. Results show that mental health services were among the 10 services considered to be most commonly needed by residents. Only 44% of residents were receiving 1 or more mental health services. High levels of unmet need were also recorded for social activities and day-care provision. 47% of residents were found to require supportive accommodation.i

i. McGilloway S, Donnelly M. Service needs of the homeless mentally ill in Belfast. International Journal of Mental Health 2001; 30(3): 50-56
(Type IV evidence – survey study of 164 homeless adults (from 27 hostels and 6 B&B houses that accommodate homeless people in Belfast) with mental health problems (83% male; mean age 41 years). A 4-section structured Needs Assessment Interview (NAI) schedule was completed for each resident by one of the mental health professionals in conjunction with one or more hostel or B&B staff.)



Ethnic minority groups

5.2g Policies to guide practice in the key areas of recording and monitoring ethnicity, dealing with racial harassment, staff training and the use of interpreters are not universally in place. The largest ethnic-minority group among the detained patients on the day of the visit was black Caribbean, comprising 42% of the total. Only half the units visited had written policies on recording patients’ ethnicity. A third of managers used ethnicitiy data to identify gaps in services and just under half used it in service planning and development. 14 units did nothing beyond recording patients’ ethnicity. Almost three-quarters of the units had no policy on dealing with racial harassment of black and ethnic-minority patients by other patients or by staff. 59 of the patients (11%) whose notes were examined had reported incidents of racial harassment.

Staff need appropriate training and information to equip them to provide mental healthcare that is senstive to the needs of all their patients. Two-thirds of the units had no training policy on race equality and anti-discriminatory practice relating to patients. Half-the units had a policy on the provision and use of interpreters and three-quarters used trained interpreters. However, two-thirds of the wards had used patients’ relatives or friends to interpret: this is of concern, as widespread use of family members as interpreters can comprosmise objective decision-making by staff. 1 in 5 of the selected patients did not have English as their first language and between them they spoke 26 languages. The proportion who were fluent in English ranged from 98% of the black Caribbean patients to only half of the Bangladeshi patients. Information on dialects spoken was poorly recorded, which has significant implications for planning care and accessing interpreters. Most ward managers said patients could request an interpreter but only 31 of the 56 patients who were not fluent in English had ever used an interpreter.i

Caveat: Interview and data collection methods have not been reported.

i. Warner L, Harris J, Ford R, Patel K, Nicholas S. Facing hard truths. Health Service Journal 2000; 110: 30-31
(Type IV evidence – observational survey of 104 inpatient units across England and Wales. Unit (n=104) and ward (n=119) managers were interviewed and case notes of detained patients (n=534; 373 male, 161 female) from black and ethnic-minority groups were examined. Current practice was explored through an unannounced interview with ward managers and by examining selected case notes.)


5.2h Results indicate that people from black ethnic minorities may make little use of the voluntary sector services because of lack of knowledge of their existence and because of invisible cultural barriers. The 6 voluntary sector providers of mental health included in the study were not meeting the needs of black ethnic minority communities because they did not know what those needs were. Without positive action the situation seems unlikely to change and there is a risk that people who may benefit from services are being excluded.i

Caveat: Participant demographic characteristics, sampling strategy, method of data collection and analysis have not been reported

i. Lai C. Reaching out to black ethnic minorities: a voluntary sector perspective on mental health. Practice 2000; 12(1): 17-28
(Type IV evidence – qualitative study. Semi-structured interviews were conducted with 6 managers, 6 workers and 7 service users from 6 voluntary sector providers of mental health in Aberdeen.)



5.2i Some of the challenging issues raised by the preliminary analysis included the need to look at the provision of adequate space and privacy for Gypsy and Traveller families on sites and the development of work oportunities for both men and women which could fit with their skills and family responsibilities. There was a need to consider family-oriented care in mental health where individuals were not removed from their carers to be ‘treated’. Many of the Gypsy and Traveller women only gained support from their family of origin, whom they often did not live near as they had joined their husband’s family. Psychological and psychiatric ‘treatments’ would appear to have a limited role in addressing the levels and complexity of difficulties experienced by this group which were contributing to their high levels of reported mental health problems. There was a clear reluctance to engage with secondary mental health services. Participants adopted strategies to maximinse the resources that they had access to. It is clear that any significant improvement in their lives will need to address the effect of distal powers over which they have no control which results in deprivation and thus distress in their lives.i

Caveat: The sampling strategy, interview methods, data collection and analysis have not been reported and sequences from the original data have not been presented. Focus group methods and details of attendees are not provided.

i. Appleton L, Hagan T, Goward P, Repper J, Wilson R. Smail's contribution to understanding the needs of the socially excluded: the case of gypsy and traveller women. Clinical Psychology 2003; 24: 40-6
(Type IV evidence – qualitative study to apply Smail’s theoretical perspective to a case example. Interviews were conducted for 13 women in a Gypsy and Traveller group where the focus of discussion was their relationships with statutory mental health services. 4 focus groups (n=8 participants per group) were also conducted with a range of service-providers.)


Rural areas
5.2j A variety of special issues in rural areas, which make mental health service provision problematic, were identified. These related to the characteristics of the rural location and community, demands upon and availability of mental health clinicians, and the changing role of mental health services. These features, together with limited access to services by patients, neccessitate models of service delivery different from those provided in urban areas. Important features include a shift from the ‘specialist as direct provider of care’ role to one of consultation, education, and indirect service provision and the use of a variety of outreach arrangements to enable patient access to essential specialist services.i

i. Judd F, Fraser C, Grigg M et al. Rural psychiatry special issues and models of service delivery. Disease Management and Health Outcomes 2002; 10(12): 771-781
(Type V evidence – literature review to appraise the problems of delivery of services to rural areas in those countries with formal mental health services, and where the availability of psychiatrist and specialist mental health practitioners approximates that recommended by the World Health Organization (e.g. Canada, North America, Britain, Australia and New Zealand.)


National Service Framework: key action 18

Adult mental health services: a national service framework for Wales. Welsh Assembly Government 2002

Local Authorities, in conjunction with Local Health Boards and voluntary agencies, should identify how they will meet the needs of groups which have particular difficulty accessing services, such as homeless people, people from BME groups (including travellers) and people with disabilities. [Key action 18 paragraph 21.2]

How can the needs of groups with difficulty accessing services be met?

The Statements The Evidence
5.3 Improving ethnic minority groups access to mental health services

5.3a The literature reviewed demonstrated distinctly different patterns of care with a higher proportion of the Black population admitted to psychiatric units and more likely to be located in locked wards or detained under the Mental Health Act. At the same time there is some evidence that certain types of conditions may be underestimated in the community (e.g. dementia, depression). In addition, the review highlights issues of acceptance of diagnoses; evaluation of training for mental health professions; and lack of research on the discharge process. Many access issues appear to be directly linked to communication difficulties and there is a need for development of culturally valid diagnostic tools. Communication difficulties also appear to limit access to some treatments (i.e. ‘talking therapies’). Studies of mental health advocacy appear to demonstrate benefits, although cost-effectiveness has not been assessed.i

Caveat: It is unclear if any of the included studies were randomised controlled trials, therefore the review has been graded as type III evidence.

i. Atkinson M, Clark M, Clay K, Johnson M, Owen D, Szczepura A. Systematic review of ethnicity and health service access for London. Coventry: CHESS University of Warwick, 2001
(Type III evidence – systematic narrative overview of 64 intervention and non-intervention studies of ethnicity and health services access for London. 17 papers reviewed the mental health literature and were divided into general discussions of the scale and type of problems and utilisation patterns (n=5), perceptions and barriers to use of services (n=4), alcohol and drug services (n=1), primary care (n=2), interventions to improve access/use of services (n=2), refugees (n=2) and anorexia nervosa (n=1). Literature search 1995-2000.)


5.3b There is strong evidence of variation between ethnic groups for voluntary and compulsory admissions, and some evidence of variation in pathways to specialist care. Most studies compared Black and White patients, finding higher rates of in-patient admission among Black patients. The pooled odds ratio for compulsory admission, Black patients compared with White patients, was 4.31 (95% CI, 3.33-5.58). Black patients had more complex pathways to specialist care, with some evidence of ethnic variations in primary care assessments.i, ii

i. Bhui K, Stansfeld S, Hull S, Priebe S, Mole F, Feder G. Ethnic variations in pathways to and use of specialist mental health services in the UK. British Journal of Psychiatry 2003; 182(2): 105-116
ii. Carter Y, Falshaw M, Feder G et al. Systematic reviews of access to and uptake of health services by ethnic groups: cardiovascular disease mental health. London: NHS Executive London, Research & Development, 2001

(Type IV evidence – systematic review and meta-analysis of 38 observational studies. Literature search 1983-2000.)

5.3c Ethnic diversity both in the characteristics of patients and their patterns of psychiatric care should be addressed when planning and developing services. There were significant differences between ethnic groups on most demographic variables in each of the diagnostic categories. There were variations in the level of contact with different mental health professionals. The only significant difference in the use of specific services was for those with psychotic/bipolar disorders, black Caribbean patients being more likely to be detained in hospital compulsorily.i

Caveat: This study was undertaken in a highly deprived inner city catchment area with a distinct demographic profile and service history and may have limited generalisibility to other settings.

i. Commander MJ, Odell SM, Surtees PG, Sashidharan SP. Characteristics of patients and patterns of psychiatric service use in ethnic minorities. International Journal of Social Psychiatry 2003; 49(3): 216-24
(Type IV evidence – cross-sectional data analysis of 2169 West Birmingham Health District residents (aged 16 to 64 years) using psychiatric services during a 6-month period in 1992. Demographic, clinical and service use data were collected from staff and records and compared across Black Caribbean, Indian, Pakistani, Irish and white ethnic groups for 2 broad diagnostic categories: psychotic/bipolar and depressive/neurotic disorders.)


5.3d Linguistic and conceptual problems explain Chinese women's relatively poor access to mental health services. The continuing failure to tackle systematically these communication problems through the routine provision of interpretation and advocacy services lays the health care system open to the charge of "institutional racism". Communication with health care professionals was hindered by a lack of common language and an absence of shared concepts concerning the causes and manifestations of health and illness, particularly mental health. This lack of communication resulted in delayed diagnoses, misunderstood treatment regimens and deterred women from (re-)presenting to the NHS. Among those informants interviewed, these types of problems were more acute for those women who were most marginalised from English-language culture.i

i. Green G, Bradby H, Chan A, Lee M, Eldridge K. Is the English National Health Service meeting the needs of mentally distressed Chinese women? Journal of Health Services & Research Policy 2002; 7(4): 216-21
(Type IV evidence – qualitative study. 42 Chinese women (aged 29-60 years) completed face-to-face interviews to collect retrospective accounts of their conceptualisation and management of mental ill-health or distress and how it was manifested. Participants had all consulted a general practitioner and had either experienced mental distress (n= 24) and/or had used traditional Chinese medicine (n = 25).)


5.3e The results suggest that the Punjabi Postnatal Screening Questionnaire (PPNSQ) may be useful in excluding postnatal depression (PND) but is no substitute for clinical assessment in suspected cases. The questionnaire was widely accepted by the participants, the clients, health visitors and link workers in Sheffield and Bradford. Both groups called for ratification of the PPNSQ and for a similar tool to be developed to meet the needs of other ethnic minority groups. Using a cut-off point of 11/12 for the PPNSQ, 2 of the women who scored above this level were found to be suffering from postnatal depression. 5 women who scored 11+ on the questionnaire did not satisfy the DSMIV (Mental Health Assessment) criteria for postnatal depression. The sensitivity was 66%; specificity 75%, Positive Predictive Value 29% and Negative Predictive Value 94%.i

i. Sobowale A. Postnatal depression in South Asian women A search for a culturally appropriate and accurate detection tool. Sheffield: Sheffield South West Primary Care Trust, 2002
(Type IV evidence – validation study. At 6 weeks postnatal, 100 women (mean age 26.5 years, Sheffield n=60; Bradfield n=40) completed the Punjabi Postnatal Screening Questionnaire (PPNSQ). This was immediately followed by a semi-structured interview to establish acceptability of the questionnaire and obtain basic demographic data. A sub group (n=37) was referred for independent mental health assessment, of whom 23 were assessed by the community psychiatric nurse. 2 focus groups were held (Group A, health visitors and link workers; Group B, women from the study) as part of the validation process.)


5.3f There is virtually no evidence of a relationship between client race, case manager race or client-case manager racial matching on outcomes or service use for homeless persons with serious mental illness recruited to a case management programme. Although African Americans had more severe problems on several measures and higher levels of service use at baseline, no differences in service use at 12-months or in the changes in client outcomes as measured by 9 variables were associated with the different pairings of African-American and white clients and case managers. White clients had a greater reduction in psychotic symptoms than did African-American clients, regardless of client-case manager racial pairing. No differences were found between white and African-American clients on the amount of services received over time.i

Caveat: Generalisability to the UK may be limited.

i. Chinman MJ, Rosenheck RA, Lam JA. Client-case manager racial matching in a programme for homeless persons with serious mental illness. Psychiatric Services 2000; 51(10): 1265-72
cgi/reprint/51/10/1265 [accessed 29/07/05
(Type II evidence – secondary data analysis using data for 1791 clients (mean age 38.4 years; 63.7% male; 53.1% African-American) from the first 2 cohorts of the Access to Community Care and Effective Services and Supports (ACCESS) programme who had completed 1-year follow-up interviews. 375 Case managers (mean age 37 years; 40.6% male; 36.4% African-American) treated clients in the study group. Analysis measured the effect of client and case manager race and their interaction on changes in outcomes and service use.)


5.3g Cost-utilisation for ethnic-specific services (ESS) Asian clients was higher than that for mainstream Asian clients. Better treatment outcome was found for ESS clients compared to their mainstream counterparts, even after controlling for certain demographics, pretreatment severity, diagnosis, and type of reimbursement. Moreover, there was a significant relationship between cost utilisation and outcome for ESS clients, whereas for mainstream clients, this relationship was not significant. The findings suggest that mental health services with an ethnic-specific focus provide more effective and efficient care for at least one ethnic minority group. Implications for the delivery of culturally-competent mental health services are discussed.i

Caveat: Generalisability to the UK may be limited.

i. Lau A, Zane N. Examining the effects of ethnic-specific services: an analysis of cost-utilization and treatment outcome for Asian American clients. Journal of Community Psychology 2000; 28(1): 63-77
(Type IV evidence – case series analysing data from 3178 (mean age 36.25 years; 53% female) Asian American clients receiving outpatient care from agencies in the Los Angeles County Department of Mental Health between 1993 and 1994 were analysed. 1981 received services at enthnic-specific agencies and 1197 received services at mainstream agencies. Cost-utilisation and clinical outcomes were evaluated.)


5.3h The effect of referring inpatients with serious mental illnesses to an ethnically focused unit varied by ethnic group, probably because each specialty unit functioned differently, depending on the needs of its particular patient population. Black patients had more diagnoses of psychotic disorders (15.2% versus 9.4% for schizo-affective disorder; 26.3% versus 19.8% for schizophrenia) and fewer diagnoses of affective disorders than other ethnic minorities or whites (7.7% versus 8.2%) and Latino patients had more non-specific diagnoses (33.6% versus 27.8%). Matching inpatients to ethnically focused units did not have a marked effect on patterns of diagnoses among black patients, but an association was observed for Latino patients, particularly those who had only 1 admission. No significant diagnostic differences were found between Asian patients and whites, irrespecitve of whether the Asian patients had been ethnically matched to a speciality focus unit.i

Caveat: As data for key variables were not always available for all patients, sample sizes for some results presented in this study were less than the total sample (n=5983). The study may have limited generalisability to the UK .

i. Mathews CA, Glidden D, Hargreaves WA. The effect on diagnostic rates of assigning patients to ethnically focused inpatient psychiatric units. Psychiatric Services 2002; 53(7): 823-9
cgi/reprint/53/7/823 [accessed 29/07/05
(Type IV evidence – retrospective data analysis. Administrative data for 5983 patients in San Francisco (20-80 years old), representing 10645 hospital admissions between 1989 and 1996 were analysed to assess the relationship between matching patients to ethnically focused units and the rates of major psychiatric illnesses among groups.)


5.3i Findings of this study provided valuable insights concerning Asian immigrants' hesitancy accessing and utilising mainstream mental health facilities. The many poignant personal anecdotes illustrate that the migration and adaptation processes can be painful and full of anguish. Unless their experiences are better understood and accepted, many Asian Canadians will likely remain outside of the available mainstream mental health facilities. Content analyses revealed 6 themes that defined a mental health problem: (a) feeling a lack of purpose in life, (b) feeling lonely, (c) difficulties understanding and dealing with a new environment, (d) high anxiety levels, (e) descriptions of mental health problems as somatic illnesses, and (f) perceptions of mental illness as serious and potentially not treatable. It was also found that poor English language ability and a lack of understanding of mainstream culture were major barriers to accessing mental health facilities.i

i. Li HZ, Browne AJ. Defining mental illness and accessing mental health services: Perspectives of Asian Canadians. Canadian Journal of Community Mental Health 2000; 19(1): 143-59
(Type IV evidence – qualitative study of face-to-face interviews with 60 Asian Canadians (mean age within each group ranged from 41 to 43 years, from 3 Asian Canadian groups: Chinese (n=20), Indian (n=20) and Filipino (n=20).)


5.4 Improving access and contact with mental health services for homeless people 

See also Section 3.2 and 5.2

5.4a Clients’ relationship with their case manager was significantly associated with homelessness and modestly associated with general life satisfaction. At 3 months, clients who had formed an alliance with their case manager had significantly fewer days of homelessness at 12 months. Clients who reported a high alliance with their case manager at 12 months had significantly fewer days of homelessness at 12 months than those with a low-alliance (6.26 days homeless in the past 60 days, compared with 8.88 days, respectively) and those with a low alliance at 12 months had fewer days of homelessness than clients who reported no relationship with their case manager (8.88 days in the past 60 days, compared with 12.54 days, respectively). Clients with a higher alliance at both 3 and 12 months reported greater general life satisfaction at 12 months.i

i. Chinman MJ, Rosenheck R, Lam JA. The case management relationship and outcomes of homeless persons with serious mental illness. Psychiatric Services 2000; 51(9): 1142-7
cgi/reprint/51/9/1142 [accessed 29/07/05
(Type II evidence – secondary analysis using data from the first 2 cohorts of the Access to Community Care and Effective Services and Supports (ACCESS) programme in America. Clinical outcomes were analysed for 2892 homeless persons (mean age 38.4 years) with serious mental illness and multivariate analyses of covariance were conducted for 2798 clients who had outcome data at 12 months.)


5.4b Homeless clients who have severe mental illness can be selectively discharged or transferred from assertive community treatment (ACT) to other services without subsequent loss of gains in mental health status, substance abuse, housing, or employment. Altogether, of clients who participated in follow-up, 8.7% participated for less than 3-months; 40.6%, for 3 to 10-months; 15.3%, for 11-13-months; and 35.3%, for 14-months or more. Controlling for potentially confounding factors, mental health, substance abuse, and housing outcomes did not significantly differ between clients who had been discharged at the time of follow-up as compared with those who had not. Those who had been discharged had worked significantly more days than those who had not (t1794 = 3.24, p<0.001), and they reported significantly less outpatient health service use though there was no decline in hospital days or receipt of public support payments.i

i. Rosenheck RA, Dennis D. Time-limited assertive community treatment for homeless persons with severe mental illness. Archives of General Psychiatry 2001; 58(11): 1073-80
(Type II evidence – secondary analysis using outcome data for 1617 homeless participants (38.6 years of age; 61.1% males, 48.3% females) with severe mental illness from the fourth annual cohort of the Access to Community Care and Effective Services and Supports (ACCESS) programme. Clients had been assessed at entry into assertive community treatment (ACT) and 18 months later. Outcomes measured overall quality of life, mental health, health and social services use and integration.)


5.4c Access to Community Care and Effective Services and Supports (ACCESS) demonstration programme was successful in terms of project-centered integration but not overall system integration. The 9 experimental sites did not demonstrate significantly greater overall systems integration than the 9 comparison sites. However, the experimental sites demonstrated better project-centered integration than the comparison sites (by wave 3, 0.65 compared with 0.57). Moreover, more extensive implementation of strategies for system change was associated with higher levels of overall systems integration (p=0.06) as well as project-centered integration (p=0.006) at both the experimental sites and the comparison sites.i

Interventions designed to increase the level of systems integration did not result in better client outcomes. Clients at all sites showed improvement in outcome measures. More extensive implementation of systems integration strategies was unrelated to these outcomes. However, clients of sites that became more integrated, regardless of the degree of implementation or whether the sites were experimental sites or comparison sites, had progressively better housing outcomes.ii

i. Morrisey JP, Calloway MO, Thakur N. Integration of service systems for homeless persons with serious mental illness through the ACCESS program. Psychiatric Services 2002; 53(8): 949-957
cgi/reprint/53/8/949 [accessed 29/07/05
(Type II evidence – follow-up evaluation study using interorganisational network data from the Access to Community Care and Effective Services and Supports (ACCESS) programme (n=947 agency respondents from 9 randomly selected experimental sites and 9 comparison sites).)
ii. Rosenheck RA, Lam J, Morrissey JP, Calloway MO, Stolar M, Randolph F. Service systems integration and outcomes for mentally ill homeless persons in the ACCESS program. Psychiatric Services 2002; 53(8): 958-966
cgi/reprint/53/8/958 [accessed 29/07/05
(Type II evidence – secondary data analysis of cohort data from the ACCESS programme. Data for 7055 homeless clients (mean age 38.4 years) with severe mental illness, were analysed for mental health symptoms, achievement of independent housing, quality of life, service use and client-level integration.)


5.4d With an appropriate service model, it is possible to engage disaffiliated populations, expand their use of human services, and improve their housing conditions, quality of life, and mental health status. Compared with persons in standard treatment members of the experimental group were more likely to attend a day programme (52.73% versus 27.30% p=0.001), had less difficulty in meeting their basic needs, spent less time on the streets (54.93% versus 28.22% reduction p=0.001), and spent more time in community housing (21% versus 9% increase p=0.025). They showed greater improvement in life satisfaction and experienced a greater reduction in psychiatric symptoms.i

Caveat: Given the difficulties of following mobile street-dwelling individuals, follow-up was low. Researchers remained in contact with 69% of subjects throughout the 24-months follow-up, but only 44% completed all 4 follow-up interviews. It is unclear if the 2 groups were similar at the start of the trial, as baseline demographic details have not been reported.

i. Shern DL, Tsemberis S, Anthony W, et al. Serving street-dwelling individuals with psychiatric disabilities: outcomes of a psychiatric rehabilitation clinical trial. American Journal of Public Health 2000; 90(12): 1873-8
(Type II evidence - randomised controlled trial of 168 street dwelling persons (mean age 40 years; 76% male) with severe mental illness allocated to receive the multi-component ‘Choices’ experimental programme (n=91) or to standard treatment (n=77) in New York City. Outcomes measured service use, quality of life, health, mental health, and social psychological status. 24 months follow-up.)


5.4e Results support the Housing First approach in reducing both homelessness and psychiatric hospitalisation for homeless individuals with mental illness. There was a significant effect of programme assignment on proportion of time homeless, with the control group spending significantly more time homeless (F[1, 195]=19.8, p<0.001) and in psychiatric hospitals (F[1,195]=7.4, p<0.01) and incurred fewer costs (F[1,173]=6.1, p<0.05) than the experimental group overall.i

Follow-up of this study has been planned to extend for a further 24-months.

Caveat: The sampling frame for this study has not been reported and the final sample size is unclear. Respondents were financially reimbursed completing interviews. It is unclear if an intention-to-treat analysis was used.

i. Gulcur L, Shinn M, Stefancic A, Tsemberis S, Fischer SN. Housing, hospitalization, and cost outcomes for homeless individuals with psychiatric disabilities participating in continuum of care and housing-first programmes. Journal of Community and Applied Social Psychology 2002; 13(2): 171-86
(Type II evidence – randomised controlled trial. 225 homeless participants in New York (mean age 41.5 years; 77% male) with a major mental illness, were allocated to receive either the experimental Housing First programme (n=126) or the control Continuum of Care programme (n=99). The experimental programme offered immediate access to independent housing without requiring psychiatric treatment or sobriety while the control programme made treatment and sobriety prerequisites for housing. 24-month follow-up.)


5.4f This mental health outreach service (MHOS) for homeless families is an innovative intervention which meets the complex and multiple needs of a vulnerable population unable to access mainstream mental health services. The primary objective of the service was to improve child mental health problems; however, the service developed in a responsive way by meeting social and practical needs of families in addition to its clinical role. Children in the experimental group had a significantly higher decrease in Strengths and Difficulties Questionnaire (SDQ) total scores (mean scores: experimental –2.64, SD 7.26; controls 1.88, SD 4.30; t =2.67, p=0.011; 95% CI –7.93 to –1.11). Having received the intervention was the strongest predictor of improvement in SDQ total scores. There was no significant impact on parental mental scores. Homeless families and staff expressed high satisfaction with the MHOS.i

i. Tischler V, Vostanis P, Bellerby T, Cumella S. Evaluation of a mental health outreach service for homeless families. Archives of Disease in Childhood 2002; 86(3): 158-63
(Type III evidence – a experimental study in the West Midlands with qualitative analysis. 27 children from 23 families received the mental health outreach service (MHOS) for homeless families and 27 children from 23 families resided in other hostels where no such service was available.)


5.4g The results suggest that the effectiveness, and ultimately the cost, of homeless services can be improved by matching the type of service to the consumer's level of psychiatric impairment and substance use rather than by treating mentally ill homeless persons as a homogeneous group. Persons with high psychiatric symptom severity and high substance use achieved better housing outcomes with the comprehensive housing programme than with case management alone. Participants with high impairment receiving case management showed an average increase of 52 days in stable housing compared with those in comprehensive housing programmes showed an average increase of 106 days in stable housing. However, persons with low and medium symptom severity and low levels of alcohol and drug use did just as well with case management alone. i

Caveat:Total follow-up at was just 58% at 12 months. For the case management only group, follow-up was 48% at 6 months and 36% at 12 months. Participants were paid $20.00 for each interview. Generalisability may be limited for a UK setting.

i. Clark C, Rich AR. Outcomes of homeless adults with mental illness in a housing program and in case management only. Psychiatric Services 2003; 54(1): 78-83
cgi/reprint/54/1/78 [01/7/05]
(Type III evidence - quasi-experimental study of 152 homeless persons with severe mental illness in Florida recruited into either a comprehensive housing programme (n=83 mean age 39.3 years; 52% male), in which consumers received guaranteed access to housing, housing support services, and case management or a programme of short-term case management only (n=69 mean age 37.8 years; 51% male). 12 months follow-up.)


5.4h This study supports the suggestion that with specialist intervention and support people with mental health problems who are homeless can live a more settled existence.There was no evidence that loss of contact with services related to age, forensic history, or substance misuse. Clients referred to the team on more than one occasion were marginally more likely to remain in contact with services. 10% of the sample had returned to the homeless circuit; 49% were resettled to more permanent accommodation. 43% were not in contact with services while 55% were still in contact 4 years later. i

i. Power C, Attenborough J. Up from the streets: a follow-up study of people referred to a specialist team for the homeless mentally ill. Journal of Mental Health 2003; 12(1): 41-49
(Type IV evidence – a retrospective case note review of a consecutive series of 100 homeless people (aged 17 to 71 years, 90% male) referred to the South Thames Assessment Resource and Training (START) team between January 1994 and December 1994. Subjects were followed up 4 years after initial contact.)




Contents Home

Health Evidence Bulletins Wales, Support Unit for Research Evidence, Cardiff University.E-mail: