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 7. Client Assessment And Care Pathways

Effective high quality care based on the best evidence and including provision for the medical, physical, psychological and social needs of service users and carers.

National Service Framework: key action 29

Raising the standard. Cardiff: Welsh Assembly Government, October 2005

The Care Programme Approach (CPA) has been introduced across Wales for all cases with a serious mental illness and/or complex enduring needs. CPA combines Care Planning and Case Management and is integrated with the Unified Assessment Process (UAP) to provide a framework for care co-ordination in mental health care. [Key action 29 paragraph 28.1]

What evidence is available to support the use of Care Programme Approach and Case Management? 

See also Section 7.16 for information about the Care Programme Approach in prisons

The Statements The Evidence
7.1 Care Programme Approach

7.1a Policy Implementation guidance issued by the Welsh Assembly Government is available regarding the Care Programme Approach. Areas covered include:

  • Making an assessment
  • Assessment of risk
  • Unmet needs
  • The care plan
  • Contingency and crisis planning
  • Support for carers
  • Role of the care co-ordinator
  • Monitoring and review
  • Loss of contact with services – enhanced CPA
  • Refusal to maintain contact – enhanced CPA
  • Confidentiality

i. Mental Health Policy Wales Implementation Guidance. The Care Programme Approach for Mental Health Service Users. Welsh Assembly Government. February 2003  [accessed 1/11/05]
(Type V evidence – expert opinion)


7.1b Introduction of clinical case management through the Care Programme Approach (CPA) was associated with an increasing focus on patients with the most severe disorders. The number of patients in contact with the service increased from 293 to 334 with an increased proportion with severe mental disorder (psychotic n=83, 24.9%; and mood disorders n=178, 53.3%) but hospitalisation did not increase. Full multidisciplinary CPA was used for patients with severe disorders and low levels of functioning.i

i. Cornwall PL, Gorman B, Carlisle J, Pope M. Ten years in the life of a community mental health team: the impact of the care programme approach in the UK. Journal of Mental Health 2001; 10(4): 441-447
(Type IV evidence – cross-sectional survey to evaluate the changes in the operation of a community mental health service in England 5-years after the implementation of the Care Programme Approach. 17 community mental health team workers completed questionnaires for their clients. Results were compared to 1992 data.)


7.1c As user involvement is an integral part of Care Programme Approach (CPA), it is important that we develop strategies that allow their views to shape, in a genuine way, the services being put in place to meet their needs. In Ayrshire they seem to have learned that their participation mattered. 4 major themes emerged from the service users: the power of user involvement, how receiving CPA can help to avert potential problems, the rights of service users, and the benefits of advocacy. These service users felt that CPA had made a real difference to their lives.i

Caveat: A potential bias may have been introduced from gathering data from such a small group of service users and from the involvement of the CPA Coordinator.

See Sections 2.42.5 for user involement in mental health services

i. Alexander H, Brady L. What does receiving the care programme approach mean for service users? Health Bulletin 2001; 59(6): 412-416
(Type IV evidence – qualitative study. 6 people with severe and enduring mental illness, receiving the Care Programme Approach in Glasgow, participated in a peer-group discussion. Service users were invited to bring a carer or friend and the group also included a representative from advocacy services.)



7.1d The evaluation demonstrated that Redford Lodge has successfully integrated risk assessment within the Care Programme Approach (CPA) process and has developed tools that offer a basis for guiding interventions while the service user is detained in hospital and to inform future strategies for supporting them in the community. Redford Lodge is to further develop its risk assessment process. Particular issues to be addressed are: streamlining the risk assessment process to reduce the clerical burden on staff and the number of duplicated records; developing the use of standardised risk assessment scales in the Redford Lodge procedures; extending the use of audit to ensure risk information is regularly updated; and monitoring the format of CPA review meetings to ensure that the discussion of risk received due consideration.i

Caveat: The response rate of external clinicians was only 45%. It is not reported how many questionnaires were sent to referring agencies at phase 2.

i. Vick N, Birke S, McKenzie R. Risk assessment and the Care Programme Approach: an independent sector initiative. British Journal of Forensic Practice 2002; 4(2): 11-18
(Type IV evidence – 2-phase qualitative study at an independent sector provider of medium secure and forensic rehabilitation psychiatric services. At phase 1, individual structured interviews and group discussions were conducted with 33 staff members and 9 representatives of referring agencies. Observation of seven CPA meetings was also conducted. Interviews and discussions were audiotaped and transcribed. Similar methods were employed at phase 2. Questionnaires were sent at random to external clinicians (n=20) and representatives of referring agencies. Phase 2 also included analysis of care plans (n=9).)

7.2 Case management

7.2a Relatives of patients receiving Intensive Case Management (ICM) did not appraise caregiving less negatively or experience less psychological distress than relatives of patients who were receiving Standard Case Management (SCM). Considerably more relatives of patients receiving ICM had contact with a case manager during the study period than relatives of patients receiving SCM (70% versus 45%).i

Intensive Case Management  appears to be a cost-effective strategy for a subgroup of patients with severe psychosis with cognitive deficits. ICM was significantly more beneficial for borderline-IQ patients than those of normal IQ in terms of reductions in days spent in hospital, hospital admissions, total costs and needs and increased satisfaction.ii

Contact frequency was more than doubled in the intensive case management group. There were proportionately more failed contacts and carer contacts but there was no difference in the average length of individual contacts or the proportion of contacts in the patients' homes.iii

Quality of Life outcome did not differ significantly by case management treatment conditions or by diagnosis and significant improvements in Quality of life (QOL) over the 2-years were observed. A better outcome was associated with improvements in depression and with the location (site) of treatment.iv

Case-load reduction is not in itself enough to reduce the need for hospital care in psychosis. Intensive case management patients spent a mean of 9.4 days less in hospital, but with a 95% CI extending from 22.1 days less to 3.2 days more. Identifying the optimal clinical profile for patients likely to benefit from intensive case management remains a pressing need for further studies. Overall reduced case-load size did not reduce hospitalisation or treatment costs over 2-years despite elimination of outliers. Age, previous hospitalisation and source of recruitment to the study all correlated with outcome.v

Intensive case management does not appear to influence the prevalence of suicidal behaviour in chronic psychosis. Predictors identified in this study confirm some previous findings.There was no significant difference in prevalence of suicidal behaviour between treatment groups. Recent attempts at suicide and multiple recent hospital admissions best predicted future

See Sections 7.197.21 for suicide prevention

i. Harvey K, Burns T, Fiander M, Huxley P, Manley C, Fahy T. The effect of intensive case management on the relatives of patients with severe mental illness. Psychiatric Services 2002; 53(12): 1580-1585
cgi/reprint/53/12/1580 [accessed 29/07/05
(Type II evidence – secondary analysis of prospective data over a 2-year period from patients participating in the UK700 trial. The trial allocated patients with severe psychotic illness to receive intensive or standard case management. At 2-years follow-up, relatives of 116 patients were interviewed.)
ii. Hassiotis A, Ukoumunne OC, Byford S, et al. Intellectual functioning and outcome of patients with severe psychotic illness randomised to intensive case management. British Journal of Psychiatry 2001; 178: 166-171
(Type II evidence – secondary analysis of 104 patients with borderline IQ and 482 patients with normal IQ, participating in the UK700 trial.)
iii. Burns T, Fiander M, Kent A, et al. Effects of case-load size on the process of care of patients with severe psychotic illness. Report from the UK700 trial. British Journal of Psychiatry 2000; 177: 427-433
(Type II evidence – secondary data anaylsis comparing patient contacts and all other patient-centred interventions of over 15 minutes for 545 patients enrolled in the UK700 trial.)
iv. Huxley P, Evans S, Burns T, Fahy T, Green J. Quality of life outcome in a randomized controlled trial of case management. Social Psychiatry & Psychiatric Epidemiology 2001; 36: 249-255
(Type II evidence – secondary analysis of quality of life outcomes at 2-year follow-up for 682 patients from the UK700 trial.)
v. Burns T, White I, Byford S, Fiander M, Creed F, Fahy T. Exposure to case management: relationships to patient characteristics and outcome. Report from the UK700 trial. British Journal of Psychiatry 2002; 181(3): 236-241
(Type II evidence – secondary analysis of hospitalisation data of 679 patients enrolled in the UK700 trial including 595 patients who were exposed for at least 50% of the follow-up period.)
vi. Walsh E, Harvey K, White I, Higgitt A, Fraser J, Murray R. Suicidal behaviour in psychosis: prevalence and predictors from a randomised controlled trial of case management: report from the UK700 trial. British Journal of Psychiatry 2001; 178: 255-260

(Type II evidence – secondary analysis assessing prevalence of suicidal behaviour in 663 patients from the UK700 case management trial.)



7.2b Case management ensures that more people remain in contact with psychiatric services (one extra person remains in contact for every 15 people who receive case management), but it also increases hospital admission rates. Case management increased the numbers remaining in contact with services (OR 0.70, 99%CI 0.50-0.98). Case management approximately doubled the numbers admitted to psychiatric hospital (OR 1.84, 99% CI 1.33-2.57). Whilst there is some evidence that case management improves compliance, it does not produce clinically significant improvement in mental state, social functioning, or quality of life. Present evidence suggests that case management increases health care costs, perhaps substantially, although this is not certain.i

Caveat: The literature search was conducted in 1997, and therefore the included trials are relatively old.

i. Marshall M, Gray A, Lockwood A, Green R. Case management for people with severe mental disorders. The Cochrane Database of Systematic Reviews 1998, Issue 2.
cochrane/clsysrev/articles/CD000050/frame.html [accessed 29/07/05
(Type I evidence – systematic review of 11 randomised controlled trials to determine the effects of case management as an approach to caring for severely mentally ill people in the community. Data were analysed for numbers remaining in contact with psychiatric services, extent of hospital admission, clinical and social outcomes, and cost. Literature search to 1997.)


7.2c Results indicate that both assertive community treatment (ACT) and clinical case management programmes (CCM) were more effective than usual treatment in 3 outcome domains: family burden, family satisfaction with services, and cost of care. The total number of admissions and the proportion of clients hospitalised were reduced in ACT and increased in CCM programmes. In both programmes, the number of hospital days used was reduced, but ACT was significantly more effective. The 2 types of case management were equally effective in reducing symptoms, increasing clients' contacts with services, reducing dropout rates, improving social functioning, and increasing clients' satisfaction.i

Caveat: Only English language papers were included and some evidence of publication bias was found.

i. Ziguras S J, Stuart G W. A meta-analysis of the effectiveness of mental health case management over 20 years. Psychiatric Services. 2000; 51(11): 1410-1421
cgi/reprint/51/11/1410 [accessed 29/07/05
(Type I evidence – systematic review of 35 studies (including 29 randomised controlled trials) comparing assertive community treatment (ACT) or clinical case management (CCM) with usual care, and 9 studies comparing ACT directly with CCM (including 7 RCTs). Literature search 1980 to 1998.)


7.2d Nurse-led case management did not result in a significant reduction in the readmission rate overall. The readmission rate in the intervention group was 9% compared to 10% in the comparison group. A history of contact with the psychiatric services emerged as an independent predictor of readmission (adjusted OR 3.05, 95% CI 1.48-6.28, p=0.0025). This was followed by a history of self harm prior to the index episode (adjusted OR 2.54, 95% CI 1.27-5.07, p=0.0084). In respect of multiple readmission, chronic alcohol problems were an independent predictor (adjusted OR 4.39, 95% CI 1.17-66.53, p=0.0288).i

i. Clarke T, Baker P, Watts CJ, Williams K, Feldman RA, Sherr L. Self-harm in adults: a randomised controlled trial of nurse-led case management versus routine care only. Journal of Mental Health 2002; 11(2): 167-176
(Type II evidence – randomised controlled trial of 467 patients (mean age 33 years, 56% female) presenting following deliberate self-harm, allocated to receive routine management enhanced by nurse-led case management or routine management only.)



7.2e Continued specialist assertive outreach service models have higher costs than non-specialist services for no apparent benefit. In the long term new assertive outreach services should have procedures in place to transfer people to lower intensity and lower cost care. All 120 live participants were traced. Only four people had no service contact; when contacted by a researcher they appeared to be coping well. No incidents of serious violence were discovered. No differences existed between teams in the mean total symptom or total social functioning change scores at follow-up, after controlling for baseline differences. No differences existed in mean cost between teams during the first 18 months. Mean (SD) annualised costs varied considerably in the 18-60 month period: sustained team £13,734 (10,820); integrated team £11,037 (13,603); disbanded team £5,742 (7,007) (F=4.4, 105 df, p=0.015).i

Caveat: During the follow-up study from month 18 to month 60, the services configuration changed; two teams were disbanded or amalgamated with community local psychiatric services.

i. Ford R, Barnes A, Davies R, Chalmers C, Hardy O, Muijen M. Maintaining contact with people with severe mental illness: 5-year follow-up of assertive outreach. Social Psychiatry & Psychiatric Epidemiology 2001; 36(9): 444-447
(Type IV evidence – multi-centred 5-year follow-up cohort study of 131 patients with severe mental illness based at 3 Intensive Care Management (ICM) teams in England practicising assertive outreach. Cost-effectiveness analysis was performed to examine whether different models of care ICM or standard community care (SCC) have differential rates of engagement, clinical outcomes, service use and costs.)


7.2f Admission to a case management service resulted in a substantial reduction in use of psychiatric inpatient care (on average reduced by 43% in terms of bed days used), which to some extent was related to specific activities of the case manager. The reduction in psychiatric inpatient care was largest for individuals with a diagnosis of schizophrenia. Case manager interventions directed towards the clients finances and coordination of care and support was related to less use of psychiatric inpatient services. More contacts with the case manager was related to fewer visits in psychiatric outpatient care. The use of primary health care and other somatic health care was unaffected.i

Several types of intervention were related to client outcome. Brokerage, intervention planning and more interventions in the area of skills relating to activities of daily living were related to a more pronounced decrease in needs of care. More time spent on indirect work on behalf of the clients was related to a better outcome with regard to psychiatric symptoms and social network.ii

i. Bjorkman T, Hansson L. How does case management for long-term mentally ill individuals affect their use of health care services?: an 18-month follow-up of 10 Swedish case management services. Nordic Journal of Psychiatry 2000; 54(6): 441-447
ii. Bjorkman T, Hansson L. What do case managers do? An investigation of case manager interventions and their relationship to client outcome. Social Psychiatry & Psychiatric Epidemiology 2000; 35(1): 43-50

(Type IV evidence – observational multi-centre study as part of an evaluation of 10 Swedish case management services. 176 long-term mentally ill individuals (mean age 41 years; 53% male) followed-up during the 18-month period before and after their registered admission. Case managers also made weekly reports during the 18-month follow-up regarding number of client contacts, type of interventions, time spent with the client in a schedule and life area of the intervention.)

National Service Framework: key action 31

Raising the standard. Cardiff: Welsh Assembly Government, October 2005

All service providers are to review annually their risk management strategies in the light of any lessons learnt or information generated by the CPA, untoward incidents and complaints. Such reviews should inform Clinical Governance and Best Value and where required have an identifiable action plan to address any isues raised. [Key action 31 paragraph 28.5]

What evidence is available regarding the assessment and management of risk to self or others? 

See Sections 7.1 for CPA and risk management
See also Sections 7.20-7.21 for information regarding suicide risk

The Statements The Evidence
7.3 Risk management

7.3a Evidence based guidelines are available that include information on the assessment and management of risk for people with schizophrenia, eating disorders, depression or people who self-harm.

Assessment of people with eating disorders should include a comprehensive assessment of risk to self. The level of risk to the patients’ mental and physical health should be monitored as treatment progresses because it may increase – for example following weight change or at times of transition between services in cases of anorexia nervosa.i

Where a patient with depression presents considerable immediate risk to self or others, urgent referral to a specialist mental health service should be arranged.ii

All people who have self-harmed should be assessed for risk: this assessment should include identification of the main clinical and demographic features known to be associated with risk of further self-harm and/or suicide, and identification of the key psychological characteristics associated with risk, in particular depression, hopelessness and continuing suicidal intent.iii

Staff on psychiatric inpatient units should be trained in how to assess and manage potential and actual violence using de-escalation techniques, restraint, seclusion and rapid tranquilisation. Staff should also be trained to undertake cardiopulmonary resuscitation. Factors to be routingely identified, monitored and comrrected include: overcrowding, lack of privacy, lack of activities, long waiting times to see staff, poor communication between patients and staff, and weak clinical leadership.iv

i. National Institute for Clinical Excellence. Eating disorders. Core interventions in the treatment and management of anorexia nervosa, bulimia nervosa and related eating disorders. Clinical guideline 9. London: NICE. January 2004. Review date: January 2008
cg009niceguidance.pdf [accessed 29/07/05]
(Evidence based guideline with systematic literature search and expert consensus.)

ii. National Institute for Clinical Excellence. Depression. Management of depression in primary and secondary care. Clinical guideline 23. London: NICE. December 2004. Review date: December 2008
[accessed 29/07/05]
(Evidence based guideline with systematic literature search and expert consensus.)
iii. National Institute for Clinical Excellence. Self-harm. The short-term physical and psychological management and secondary prevention of self-harm in primary and secondary care. Clinical Guideline 16 London: NICE July 2004. Review date: July 2008
CG016NICEguideline.pdf [accessed 29/07/05]
(Evidence based guideline with systematic literature search and expert consensus.)
iv. National Institute for Clinical Evidence. Schizophrenia. Core interventions in the treatment and management of schizophrenia in primary and secondary care. Clinical guideline 1. London: NICE. December 2002. Review date: December 2006 [accessed 29/07/05]

(Evidence based guideline with systematic literature search and expert consensus.)



7.3b A NICE evidence based guideline on the management of disturbed behaviour in in-patient psychiatric settings and when service users present for mental health assessment in emergency departments, is now available. The following interventions and related topics are covered in this guideline:

  • Environment, organisation and alarm systems

  • Prediction (antecedents, warning signs and risk assessment)

  • Training

  • Service user perspectives, including those relating to ethnicity, gender and other special concerns

  •  Searching

  • De-escalation techniques

  • Observation

  • Physical intervention

  • Seclusion

  • Rapid tranquillisation

  • Post-incident reviews

  • Emergency departments. i

i. National Institute for Clinical Evidence. Disturbed (violent) behaviour: the short-term management of disturbed (violent) behaviour in in-patient psychiatric settings and emergency departments. Clinical Guideline 25. London: NICE. February 2005.
cg025niceguideline.pdf [accessed 29/07/05

(Evidence based guideline with systematic literature search and expert consensus.)


7.3c Intensive case management does not reduce the prevalence of violence in psychotic patients in comparison with standard care. No significant reduction in violence was found in the intensive case management group compared with the control group (22.7% versus 21.9%, p=0.86).i

Caveat: Physical assault was the primary measure of violence. Other outcomes such as damage to property, threats or attempted assault were not collected.

i. Walsh E, Gilvarry C, Samele C, et al. Reducing violence in severe mental illness: randomised controlled trial of intensive case management compared with standard care. British Medical Journal 2001; 323(7321): 1093-1096
reprint/323/7321/1093 [accessed 29/07/05
(Type II evidence – 2 year follow-up data analysis of 708 patients (mean age 38.25 years, 57% male) with psychotic illness allocated to either intensive or standard case management. Physical assault was the primary outcome.)

7.3d A total of 138 discharged psychiatric patients (89%) had at least one post-baseline assessment and of these patients, 16 (12%) had at least one police contact in the year of the study, most of which were emergency assessments. The data showed significantly greater numbers of police contacts in patients with increasing severity of personality disturbance. Patients with such disturbance were six times more likely to have police contacts than those with no personality disorder. There were significantly more contacts in patients with borderline and antisocial (dissocial) personality disorder allocated to community-oriented care compared with hospital-oriented care. These findings have important implications for risk assessment in severe mental illness.i

Caveat: Sample characteristics have not been reported. It is unclear if an intention-to-treat analysis was used.

i. Gandhi N, Tyrer P, Evans K, McGee A, Lamont A, Harrison-Read P. A randomized controlled trial of community-oriented and hospital-oriented care for discharged psychiatric patients: influence of personality disorder on police contacts. Journal of Personal Disorders 2001; 15(1): 94-102
(Type II evidence –randomised controlled trial of 155 patients with serious mental disorders in
England allocated to either a community multidisciplinary team or to a hospital-based care programme after discharge from patient care. Assessment of personality, police contacts and costs were analysed at 1-year follow-up.)



7.3e The results of the aggregated meta-analysed studies assessing the Assaulted Staff Action Programme (ASAP) intervention suggest ASAP may be a powerful intervention in reducing frequency of assaultive behaviour. This beneficial effect was obtained across a wide range of subject groups in a wide variety of patient care units in both inpatient and community settings. Results yielded a highly statistically significant Cohen's d of 3.1 and fail-safe number of 202.i

Caveat: All of the studies analysed were conducted by the primary author who designed the ASAP. The results of this analysis may not be generalisable to a UK setting.

i. Flannery RB Jr, Everly GS Jr, Eyler V. The assaulted staff action program (ASAP) and declines in assaults: a meta-analysis. International Journal of Emergency Mental Health 2000; 2(3): 143-148
(Type III evidence – meta-analysis of 5 empirical investigations to determine the effectiveness of the Assaulted Staff Action Program (ASAP) in reducing frequency of assault. Effect sizes across studies were combined using the unweighted meta-analytic statistical approach.)


7.3f Assessment for risk of harm to others is not a part of the emergency consultation that is emphasised by the majority of junior psychiatrists. Changing practice will require a shift in the way that risk to others is presented in psychiatric teaching. Risk factors were recorded more frequently for harm to self than for harm to others. There was little recorded evidence that consideration had been given to the overall risk of harm to self, and there was no evidence of this for harm to others. Recording of risk did not change significantly between 1999 (pre-intervention) and 2000 (post-intervention). i

Caveat: The samples of patient records analysed are small. Information on whether all the junior doctors attended the intervention is not provided.

i. Stone J, Szmukler G. An audit of risk assessment in an emergency setting. Psychiatric Bulletin. 2002; 26: 88-90
(Type III evidence – before and after study. 35 patient records from the emergency clinic at the Maudley Hospital London, were analysed from July 1999 to assess the standard of risk assessment for self-harm and harm to others routinely recorded by junior doctors. The junior doctors then received an intervention about the importance of risk assessment, in 2000. A further 35 patient records were then analysed post-intervention.)


7.3g The Historical Clinical Risk Management Guide-

20 (HCR-20) and Violence Risk Scale 2 (VRS) did not predict inpatient violence within the first 6 months of admission. However, the clinical sub-scale of the HCR-20 was predictive of violence, abuse or harassment. When considering repetitiveness there was some indication across the scales that static factors predicted isolated incidents and dynamic factors repetitive violence. A number of individual items within the scales appeared to act as predictive or protective factors for inpatient violence. This study provides some indication of the differential utility of these structured clinical assessments for predicting short-term risk of violence in inpatients. In particular, the use of dynamic clinical factors in identifying those likely to engage in imminent repetitive violence.i

i. Grevatt, M, Thomas-Peter, G, Hughes, J. G. Violence, mental disorder and risk assessment; can structured clinical assessments predict the short-term risk of inpatient violence? Journal of Forensic Psychiatry and Psychology 15; 2: 278 – 292
(Type IV evidence – retrospective file review to assess the ability of the HCR-20 and VRS structured clinical risk assessments to predict violent behaviour within the first 6 months of admission to a secure forensic service, as measured by official incident reports. Information available at time of admission was used to complete the Historical and Clinical scales of the HCR-20 and VRS for a sample of 44 male inpatients.)



7.3h No controlled studies exist that evaluate the value of seclusion or restraint in those with serious mental illness. There are reports of serious adverse effects for these techniques in qualitative reviews. Alternative ways of dealing with unwanted or harmful behaviours need to be developed. Continuing use of seclusion or restraint must therefore be questioned from within well-designed and reported randomised trials that are generalisable to routine practice.i

i. Sailas E, Fenton M. Seclusion and restraint for people with serious mental illnesses. The Cochrane Database of Systematic Reviews 2000, Issue 1
clsysrev/articles/CD001163/frame.html [accessed 29/07/05
(Type I evidence – systematic review on the effects of seclusion and restraint for those with serious mental illnesses and the effectiveness of seclusion prevention strategies. Literature search to 1999.)

7.3i The absence of occupational therapy literature on the subject of risk assessment and risk management suggests that this is not usually considered part of the occupational therapists’ role. Occupational therapists, however, should maintain safe working practices and the awareness of the risk of violence from their clients is surely one of these. Clinical factors in the prediction of violence risk include accuracy of prediction, clinical versus actuarial prediction, predictability of violence, and assaults on staff. Contextual factors include social networks, conditional judgments, and neighbourhood characteristics.i

i. Blank A. Patient violence in community mental health: a review of the literature. British Journal of Occupational Therapy 2001; 64(12): 584-589
(Type V – expert opinion)


7.3j A Royal College of Psychiatrists council report on the assessment and clinical management of risk of harm to other people is currently under review, to determine whether the original report should be, reconfirmed as current policy, updated as necessary, or wthdrawn.i

i. The Royal College of Psychiatrists Special Working Party on Clinical Assessment and Management of Risk. Assessment and clinical management of risk of harm to other people. London: Royal College of Psychiatrists, April 1996: Council Report CR 53
(Type V evidence – council report of Royal College of Psychiatrists policy.)

National Service Framework: key action 32

Raising the standard. Cardiff: Welsh Assembly Government, October 2005

All users of specialist mental health services who have a serious mental illness or complex needs are to be offered written copies of their care plans drawn up in collaboration with them and their carer. This should be a holistic plan and will contain at least such details as the action to be taken in a crisis by the service user, their carer, and their care-coordinator. [Key action 32 paragraph  29.1]

What are the benefits of  patient held written care plans?

The Statements The Evidence
7.4 Written care plans

7.4a Use of joint crisis plans reduced compulsory admissions and treatment in patients with severe mental illness. The reduction in overall admission was less. This is the first structured clinical intervention that seems to reduce compulsory admission and treatment in mental health services. Use of the Mental Health Act was significantly reduced for the intervention group, 13% of whom experienced compulsory admission or treatment compared with 27% of the control group (risk ratio 0.48, 95% CI 0.24 to 0.95, p = 0.028). As a consequence, the mean number of days of detention (days spent as an inpatient while under a section of the Mental Health Act) for the whole intervention group was 14 compared with 31 for the control group (difference 16, 95% CI 0 to 36, p = 0.04). For those admitted under a section of the Mental Health Act, the number of days of detention was similar in the two groups. The intervention group had fewer admissions (30% versus 44%, risk ratio 0.69, 95% CI 0.45 to 1.04, p = 0.07). There was no evidence for differences in bed days.i

Caveat: Although the intervention group had fewer admissions the results were not statistically significant.

i. Henderson C, Flood C, Leese M, Thornicroft G, Sutherby, Szmukler. Effect of joint crisis plans on use of compulsory treatment in psychiatry: single blind randomised controlled trial. British Medical Journal 2004; 329(7428): 136 - 139
[accessed 29/07/05]
(Type II evidence - randomised controlled trial of 160 people from 8 community mental health teams in southern England, with a psychotic illness ot non-psychotic bipolar disorder and had experienced a hospital admission within the previous two years. Participants were allocated to receive joint crisis plans or a control group with information leaflets. 15 months follow-up.)



7.4b There is no quality evidence to suggest that patient-held record should be introduced as part of routine shared care for all patients with schizophrenia. However, the patient-held record was acceptable to patients with schizophrenia and acted as a communication tool, particularly between patients and keyworkers. A total of 63/92 (68.5%) patients still had the patient-held record, 64/92 (69.6%) had used it and 39 (60.9%) of the 64 who had used it said the patient-held record had no significant effect on primary outcomes (Verona Service Satisfaction Scale-54:F 1, 116=0.06, p=0.801, Krawiecka & Goldberg rating scale:F 1, 116=0.6, p=0.439) or on use of services. A higher symptom score was associated with not using the patient-held record.i

Caveat: It is unclear how many general practices were finally randomised or how many were randomised to each arm.

i. Lester HE, Allan T, Wilson S, Jowett S, Roberts L. A cluster randomised controlled trial of patient-held medical records for people with schizophrenia receiving shared care. British Journal of General Practice 2003; 53: 197-203
(Type II evidence – cluster randomised controlled trial of 201 patients (mean age 46 years; 62.1% male), from 74 general practices, with schizophrenia and receiving shared care in contact with secondary services. Allocatioin was to either a patient-held medical record intervention or a control group. Primary outcomes, at 12-months, measured, satisfaction with community mental health services and psychopathology.)


7.4c Patient-held records may not be helpful for patients with long–term mental illness. Carrying a shared care record had no significant effect on mental state or satisfaction with psychiatric services. Compared with controls, patients in the shared care group were no more likely to be admitted (RR 1.2 95% CI 0.86-1.67) and attend clinic (RR 0.96, 95%CI 0.67-1.36) over the study period. Uptake of the shared care scheme was low by patients and professionals alike. Subjects with psychotic illness were significantly less likely to use their records (RR 0.51, 95%CI 0.27-0.99).i

Caveat: The total sample size is small and the number of clients in the intervention group is greater than in the control group (n=55 versus n=35). Reported numerical results do not appear to be complete. Confidence intervals reported in the abstract do not appear elsewhere.

i. Warner JP, King M, Blizard R, McClenahan Z, Tang S. Patient-held shared care records for individuals with mental illness: randomised controlled evaluation. British Journal of Psychiatry 2000; 177: 319-324
(Type II evidence – 12-month, cluster randomised controlled trial of 90 patients with long-term mental illness (mean age 38.5 years, 53.3%) from 28 general practices in London, allocated to either a shared care record intervention or control group.)


7.5 Actions to take in a crisis
See Section 6.5 for futher information on crisis services

7.5a Within the limits of expert opinion and with the expectation that future research data will take precedence, these guidelines provide some direction for addressing common clinical dilemmas in the management of psychiatric emergencies and can be used to inform clinicians in acute care settings regarding the relative merits of various strategies.i

i. Allen MH. Treatment of behavioural emergencies: A summary of the Expert Consensus Guidelines. Journal of Psychiatric Practice 2003; 9(1):16-38
(Type V evidence – expert opinion based on a written survey of 808 decision points completed by 50 experts)


National Service Framework: key action 34

Raising the standard. Cardiff: Welsh Assembly Government, October 2005

People with mental health problems have the same needs for effective care of physical health problems including dental, visual and hearing needs as the general population.. [Key action 34 paragraph 30.1]

Primary care, working jointly with the mental health services and with the support of specialist services such as the community dentistry services are to ensure all those requiring care have access to and receive effective services, whatever their circumstances. [Key action 34 paragraph 26.1]

Are the physical health needs of people with mental health problems being met?
How can the physical health care of people with mental health problems be improved

The Statements The Evidence
7.6 Physical health needs of people with mental health problems
7.6a The assessment of needs for health and social care for people with schizophrenia should be comprehensive and address medical, social, psychological, occupational, economic, physical and cultural issues. Primary and secondary care services, in conjunction with the service user, should jointly identify which service will take the responsibility for assessing and monitoring the physical health care needs of service users. This should be documented in both primary and secondary care notes/plans and clearly recorded by care coordinators for those on the enhanced care programme approach.i

i. National Institute for Clinical Evidence. Schizophrenia. Core interventions in the treatment and management of schizophrenia in primary and secondary care. Clinical practice algorithms and pathways to care – No. 1. London: NICE, December 2002. Review date: December 2006
(Evidence based guideline with systematic literature search and expert consensus.)


7.6b Findings suggest that service users may be less likely than other vulnerable groups to receive physical health checks or to access services like smoking cessation and this is symptomatic of a range of inequalities affecting people with mental health problems. Achieving more equitable access to health promotion information, services and support should be part of a broader agenda to tackle the discrimination and exclusion experienced by people with mental health problems. For service users, a key issue was the preceived attitude and awareness of primary care staff. Myths and stereotypes about people with mental health problems were seen to influence the way GPs and other staff treat service users, and the extent to which their physical health needs are taken into account.i

Caveat: The sampling strategy, and the method of data collection and analysis have not been reported in this study.

i. Friedli L, Dardis C. Not all in the mind: mental health service user perspectives on physical health. Journal of Mental Health Promotion 2002; 1: 36-46
(Type IV evidence – qualitative study of 9 focus groups, held in day and drop-in centres and a residential house in London. 4 focus groups of 26 participants (aged 21 to 59 years) explored physical health issues in general and 5 groups with 24 participants (smokers (20+ per day); aged 21 to 59 years) looked specifically at smoking cessation.)


7.6c Despite the inextricable link between physical well-being and mental health, professionals in both primary and secondary care fail to view users holistically. Professional role ambiguity and poor communication result in access difficulties for users and add to the burden felt by carers. A focus on reactive interventions to ill-health rather than on health promotion and physical well-being took place in a context of paternalism, strict adherence to the medical paradigm and failure to take users’ physical health concerns seriously. Drug-induced weight gain was particularly distressing and had a negative impact on physical health and the desire and ability to pursue personal goals.i

Caveat: Three of the group facilitators knew some of the user and carer group and nursing staff participants, thus introducing higher potential for interviewer bias.

i. Dean J. Mum I used to be good looking...look at me now: the physical health needs of adults with mental health problems: the perspectives of users, carers and frontline staff. International Journal of ental Health Promotion 2001; 3: 16-24
(Type IV evidence – qualitative study of 4 parallel focus group interviews of users and carers in Cambridge. Structured interviews were conducted in both in- and out-patient settings with 8 nurses. Additionally, 4 staff from each of the statutory community day services were interviewed.)



7.6d Depressed patients with comorbid medical disorders tend to have similar rates of treatment but worse depression outcomes than depressed patients without comorbid medical illness. Quality improvement programmes for depression can improve treatment rates and outcomes for depressed primary care patients with comorbid medical illness. Among the depressed patients with comorbid medical disorders, the combined quality improvement programmes resulted in greater use of antidepressant medications and psychotherapy and lower rates of probable depressive disorders at both 6- and 12-month follow-up than did the usual care depression treatment programme.i

Caveat: The results of this study may have limited generalisability to the UK or to non-managed-care settings.

i. Koike AK, Unutzer J, Wells KB. Improving the care for depression in patients with comorbid medical illness. American Journal of Psychiatry 2002; 159(10): 1738-1745
(Type II evidence – secondary analysis of a cluster randomised controlled trial conducted in America. 1,356, patients, (mean age 43 years, 72.4% female), with major depression, from 46 managed primary care clinics, were allocated to either a usual care depression treatment programme or one of 2 quality improvement programmes for depression. Outcomes were assessed at 6 and 12 months.)


7.6e General Practitioner’s were aware of co-morbid psychiatric symptoms in chronic physical illness and alluded to the need for detecting and treating psychological illness appropriately. That the psychological illness may not be detected was perceived by the respondents to be due to many factors, some related to the practitioner and others to the patient. The GPs were resistant to the use of screening instruments and expressed concern about the lack of resources available to this group of patients as well as their lack of training in this area of their work.i

i. Chew-Graham CA, Hogg T. Patients with chronic physical illness and co-existing psychological morbidity: GPs' views on their role in detection and management. Primary Care Psychiatry 2002; 8(2): 35-39
(Type IV evidence – qualitative study of 13 GPs in Northwest England who completed semi-structured interviews.)


7.6f It is evident that co-morbidity of depression and physical disorders leads to problems in the recognition of depression. Depression is a condition that responds well to treatment. However, one of the main barriers to effective treatment is failure by health care professionals to identify the problem. Primary health care teams need to agree protocols and communition systems for recognising and caring for people with depression that could help address the problems of fragmentation in the primary care setting. There is also a case to be made for a much greater emphasis to be given to all aspects of mental health, in particular depression, in the education of all primary health care nurses.i

i. Martin F. Co-morbidity of depression with physical illnesses: a review of the literature. Mental Health and Learning Disabilities Care 2001; 4(12): 405-408
(Type V evidence – expert opinion.)


Dental health needs of individuals with mental health problems
7.6g Guidance and recommendations for the oral health care of people with mental health problems are available. Recommendations include client centred services, service planning and training issues.i

i. Griffiths J, Jones V, Leeman, I et al. Oral Health Care for people with Mental Health problems. Guidelines and Recommendations. British Society for Disability and Oral Health. January 2000
(Expert consensus guidelines.)

7.6h Dentists who have received an overview of some of the most significant mental illnesses to affect men, women and children – disorders collectively known as mood disorders, are better able to screen for these illnesses, consult with psychologists, psychiatrists, social workers and physicians regarding their patients with mood disorders, and understand the treatments utilised and the potential consequences for their patients.i

Caveat: The study is based on 3 case-studies in the U.S., with recommendations that may have limited generalisability in the UK.

i. Herzig BR, Belkin HR. Mood disorders in dental patients. Texas Dental Journal 2001; 118: 242-253
Type V evidence – expert opinion with 3 supporting case studies.)



7.6i The prevalence of dental disease usually is extensive because of poor oral hygiene and medication-induced xerostomia. Preventive dental education, saliva substitutes and anticaries agents are indicated. To avoid adverse drug interactions with the usually prescribed psychiatric medications, special precautions should be taken when administering certain antibiotics, analgesics and sedatives.i

i. Friedlander AH. i. Friedlander IK, Marder SR. Bipolar I disorder:psychopathology, medical management and dental implications. Journal of the American Dental Association 2002; 133: 731-740
(Type V evidence – expert opinion.)


7.7 Deaf people with mental health needs
See also Section 6.11 for specialist services for deaf people with mental health problems

7.7a Best practice guidance is available providing advice on ways to promote mental health and improve services for people who are Deaf.i

Recommendations have been made following responses to the consultation paper published in 2002, which outlined the development of a national strategy for mental health services in England, for people of all ages who are Deaf or Deafblind.ii

i. National Institute for Mental Health in England. Mental Health and Deafness. Towards Equity and Access. Deparment of Health, 2005
04/10/40/05/04104005.pdf [accessed 29/07/05
(Expert consensus guidelines.)
i. Smith J. A sign of the times: modernising mental health services for people who are deaf. London: Department of Health, 2002
04/07/67/64/04076764.pdf [accessed 29/07/05
(Type V evidence – expert opinion.)

7.7b Deaf residents with mental health probems had high levels of functional impairment, with two-thirds having moderate or severe problems in at least one domain of personal functioning including cleanliness, cooking, shopping, use of transport and budgeting (including 35% of those who had not actually received a formal diagnosis). The domains of social activity and risk of harm to self and others, more than mental illness per se, differentiated residents in psychiatric wards from those in staffed hostels. These residents are generally younger than their hearing counterparts, and appear to have had a much lower level of contact with psychiatric services, in spite of their obvious and significant mental health problems.i

i. McClelland R, Chisholm D, Powell S. Mental health and deafness: an investigation of current residential services and service users throughout the UK. Journal of Mental Health 2001; 10(6): 627-363
(Type IV evidence – cross-sectional survey to assess all residential facilities that provide servicess to deaf people with mental health problems in England, Scotland and Wales. 555 residents living in 44 residential centres of whom 372 were deaf adults aged 16-65 years were surveyed in 1996. 20 facilities with at least two residents with a mental health diagnosis and 160 of their residents were then assessed in more detail.)


7.7c Findings suggest that 60% of the sample of forensic referrals to specialist psychiatric services for deaf people, would have benefited from specialist medium secure unit services for deaf people. A high percentage of the sample had been convicted of / were currently charged with offences of violence and sexual offences, including 19.8% with sexual offences and / or offences of violence against children.

Sentencing data was available for 270 patients. 66 subjects (24.5%) had served / were serving custodial prison sentences, while 132 (subjects received probation supervision. Data was available for 179 of the 389 patients on fitness to plead, and 61 patients (34.1%) were classified as unfit to plead.i

Caveat: Some outcome measures are not available for all included patients.

i. Young A, Howarth P, Ridgeway S, Monteiro B. Forensic referrals to the three specialist psychiatric units for deaf people in the UK. Journal of Forensic Psychiatry 2001; 12(1): 19-35
(Type IV evidence – data analysis of 5034 referrals, between 1968 and 1999, to 3 specialist psychiatric services for deaf people to identify forensic cases. 431 patient files for 389 deaf individuals were identified. Data were collected on personal characteristics, offending behaviour, court disposal, and diagnosis.)



7.7d The needs of deaf people with mental health needs, who come into contact with the criminal justice system, are primarily not recognised or not met because of misdefinition, misunderstanding and inadequate or inappropriately designed responses to their offending behaviour. In important ways that are not yet fully understood, the pattern of offending behaviour and the mental health needs associated with it seem to differ from that of the hearing population. There remains a great deal of research to be carried out in order to do the most simple thing – to describe adequately this population and to provide appropriate and targeted responses to its needs.i

i. Young A, Montiero B, Ridgeway S. Deaf people with mental health needs in the criminal justice system: a review of the UK literature. The Journal of Forensic Psychiatry 2000; 11(3): 556-70
(Type V evidence – expert opinion.)


7.7e Sign language interpreters in mental health settings face extreme linguistic and cultural difficulties in interpreting basic, everyday language used in these settings. This is particularly true when deaf clients have limited English proficiency, which often requires interpreters to use expansion techniques in order to render messages successfully.i

i. Vernon M, Miller K. Interpreting in mental health settings: issues and concerns. American Annals of the Deaf 2001; 146(5): 429-434
(Type V evidence – expert opinion.)


National Service Framework: key action 36

Raising the standard. Cardiff: Welsh Assembly Government, October 2005

Primary Care Teams, Community Mental Health Teams and LHBs are to develop medicine management systems for those people where medication is part of the care plan. [Key action 36 paragraph 30.3]

What interventions help improve patients compliance with medication?
Are there any examples of previous medicine management systems that have worked well? 

See Section 6.22 for psychoeducation interventions and their effect on compliance

The Statements The Evidence
7.8 Interventions to improve management of and compliance with medication
7.8a The full benefits of medications cannot be realised at currently achievable levels of adherence. Current methods of improving adherence for chronic health problems are mostly complex and not very effective. Innovations to assist patients to follow medication prescriptions are needed. For short-term treatments, 1 of 3 interventions reported in 3 randomised controlled trials (RCTs) showed an effect on both adherence and clinical outcome. 18 of 36 interventions for long-term treatments reported in 30 RCTs were associated with improvements in adherence, but only 16 interventions led to improvements in treatment outcomes. Almost all of the interventions that were effective for long-term care were complex, including combinations of more convenient care, information, reminders, self-monitoring, reinforcement, counselling, family therapy, and other forms of additional supervision or attention by a health care provider. Two studies showed that telling patients about adverse effects of treatment did not affect their adherence.i

i. Haynes RB, McDonald H, Garg AX, Montague P. Interventions for helping patients to follow prescriptions for medications. The Cochrane Database of Systematic Reviews 2002, Issue 2
cochrane/clsysrev/articles/CD000011/frame.html [accessed 29/07/05
(Type I evidence – systematic review of 33 randomised controlled trials of patients who were prescribed medication for a medical (including psychiatric disorder). Outcome measures included medication adherence and treatment effect. Literature search to 2001.)



7.8b Psychoeducational interventions without accompanying behavioural components and supportive services are not likely to be effective in improving medication adherence in schizophrenia. Models of community care such as assertive community treatment and interventions based on principles of motivational interviewing are promising. Providing patients with concrete instructions and problem-solving strategies, such as reminders, self-monitoring tools, cues, and reinforcements, is useful. Problems in adherence are recurring, and booster sessions are needed to reinforce and consolidate gains. Thirteen (33%) of 39 identified studies reported significant intervention effects. Interventions targeted specifically to problems of nonadherence were more likely to be effective (55%) than were more broadly based treatment interventions (26%). One-half of the successful interventions not specifically focused on nonadherence involved an array of supportive and rehabilitative community-based services.i

A Cochrane systematic review to assess the effect of 'compliance therapy' on adherence with antipsychotic medication in people with schizophrenia or related psychoses compared to treatment as usual is currently underway.ii

i. Zygmunt A, Olfson M, Boyer CA, Mechanic D. Interventions to improve medication adherence in schizophrenia. American Journal of Psychiatry 2002; 159(10): 1653-1664
(Type I evidence – narrative systematic review of 33 randomised controlled trials and 6 quasi-experimental studies. Literature search to 2000.)

ii. McIntosh A, Conlon L, Lawrie S. Compliance therapy for schizophrenia. The Cochrane Database of Systematic Reviews 2004, Issue 4
cochrane/clsysrev/articles/CD003442/frame.html [accessed 29/07/05]


7.8c Medication compliance can be improved by certain, sometimes complex interventions; however further efforts are needed in developing effective interventions to assist patients in following prescribed treatment regimes. 6 studies showed improvement in compliance rates following interventions, although only 3 of these reached statistical significance. Effective interventions included individualised behaviour tailoring regimes and compliance therapy.i

i. Dodds F, Rebair-Brown A, Parsons S. A systematic review of randomized controlled trials that attempt to identify interventions that improve patient compliance with prescribed antipsychotic medication. Cinical Effectiveness in Nursing 2000; 4: 47-53
(Type I evidence – narrative systematic review of 8 randomised controlled trials. Literature search to 1999.)


7.8d A systematic review of medication adherence-enhancing interventions for patients with depression found no consistent indications of which may be effective. Carefully designed clinical trials are needed to clarify the effect of single and combined interventions. Patient education and medication clinics were the interventions most commonly tested, combined with a variety of other interventions.i

Caveat: Unpublished research was not sought.

i. Pampallona S, Bollini P, Tibaldi G, Kupelnick B, Munizza C. Patient adherence in the treatment of depression. British Journal of Psychiatry 2002; 180: 104-109
(Type I evidence – narrative systematic review of 32 studies (including 14 randomised controlled trials) to review factors associated with adherence and adherence-enhancing interventions. Literature search from Januray 1990 to December 1999.)


7.8e Compliance with once-weekly fluoxetine was better than that with once-daily fluoxetine. Compliance decreased over time when patients remained on daily dosing; however, when patients switched from daily dosing to weekly dosing, compliance did not decrease. For those patients randomly assigned to weekly fluoxetine, compliance was 85.4% during study period 1 while on treatment with daily fluoxetine and then 87.5% while on treatment with weekly fluoxetine. This difference was not significant. For once-daily dosing, however, compliance declined from 87.3% during period 1 to 79.4% during period II (p < 0.001). After adjusting for compliance during study period 1, weekly compliance during study period II was 87.8% and daily compliance was 79.0%, a statistically significant difference (p = 0.006).i

Caveat: It is unclear whether or not an intention-to-treat analysis was used.

i. de Klerk E. Patient compliance with enteric-coated weekly fluoxetine during continuation treatment of major depressive disorder. Journal of Clinical Psychiatry 2001; 62(Suppl 22): 43-47
(Type II evidence – randomised controlled trial 109 patients from the UK (mean age 46 years; 83% female) who had responded to fluoxetine treatment for 6-16 weeks. Participants were allocated to either once-weekly 90mg/week or once-daily 20 mg/day fluoxetine for 12 weeks.)



7.8f The addition of time-limited behavioural family therapy to monthly support groups improved family atmosphere, but did not influence patient social functioning or family burden. Applied Family Management (AFM) was associated with lower rejecting attitudes by relatives toward patients and less friction in the family perceived by patients. Patients in both AFM and Supportive Family Management (SFM) improved in social functioning but did not differ, whereas family burden was unchanged. Medication strategy had few effects, nor did it interact with family intervention.i

See Section 6.17 for family interventions

i. Mueser KT, Sengupta A, Schooler NR, et al. Family treatment and medication dosage reduction in schizophrenia: effects on patient social functioning, family attitudes, and burden. Journal of Consulting and Clinical Psychology 2001; 69(1): 3-12
(Type II evidence – double-blind randomised controlled trial of 313 patients with schizophrenia from 5 hospitals in the US. Patients were allocated to either supportive family management (mean age 28.8 years; 65% male) or applied family management (mean age 30.3 years, 67% male). 2 year follow-up.)


7.8g Expressed Emotion (EE) may be an important factor to account for in the understanding of patients' compliance and the direction of the relationship between EE and compliance should be the subject of further study. A number of factors were related to compliance, including carers' EE (association with compliance category OR =12.5, p=0.04) and patients' psychotic symptoms (OR =0.89, p=0.05) which contributed independently to not taking medication. Carers' knowledge about schizophrenia and other groups of symptoms was not related to compliance.i

i. Sellwood W, Tarrier N, Quinn J, et al. The family and compliance in schizophrenia: the influence of clinical variables, relatives’ knowledge and expressed emotion. Psychological Medicine 2003; 33: 91-96
(Type II evidence - study implemented during a randomised controlled family intervention trial in Manchester involving 79 patient-carer pairs randomised to a needs-based cognitive-behavioural family intervention or control group.)


7.8h Sustained periods of outpatient commitment may significantly improve adherence with community-based mental health treatment for persons with severe mental illness and thus may help improve other clinical outcomes affected by adherence. Randomised control and outpatient commitment groups did not differ significantly in group comparisons of treatment adherence. However, analyses of all subjects, including nonrandomised violent subjects, showed that those who underwent sustained periods of outpatient commitment (6-months or more) were significantly more likely to remain adherent with medication and other treatment, compared with those who underwent only brief outpatient commitment or none. Administration of depot antipsychotics also significantly improved treatment adherence independently of the effect of sustained outpatient commitment.i

Caveat: The sample included a subgroup of subjects with a recent history of serious violent behaviour who could not be randomly assigned to the initial control group and it is unclear how many subjects were randomly assigned at baseline. This US study may have limited applicability to the UK.

i. Swartz MS, Swanson JW, Wagner HR, Burns BJ, Hiday VA. Effects of involuntary outpatient commitment and depot antipsychotics on treatment adherence in persons with severe mental illness. Journal of Nervous & Mental Disease 2001; 189(9): 583-592
(Type II evidence – randomised controlled trial of 331 involuntarily hospitalised subjects from North Carolina, with severe mental illness awaiting discharge under outpatient commitment (mean age 39 years, 53.4% male). Subjects were assigned to be released or continue under outpatient commitment after hospital discharge. A nonrandomised group with a recent history of serious violence was also studied under outpatient commitment. 12 month follow-up.)


7.8i Pharmacist-run services may be accompanied by improvements in clinical outcomes. Inconsistent definitions used in the research evaluated meant that an overall interpretation of a change in the incidence of compliance and adverse drug reactions was impossible. Other outcomes such as knowledge and satisfaction showed equivocal results overall. There was little or no change in quality of life where this was assessed. Savings in drug acquisition costs may have accrued, but it was impossible to calculate the magnitude. Pharmacist involvement produced a positive impact on cost-benefit and cost-effectiveness.i

Caveat: Only 3 studies (before and after designs) investigated pharmacist reviewing for patients requiring psychiatric drug treatment. Only English language published papers were included in the review.

i. Tully MP, Seston EM. Impact of pharmacists providing a prescription review and monitoring service in ambulatory care or community practice. The Annals of Pharmacotherapy 2000; 34: 1320-31
(Type I evidence – systematic review of  13 randomised and 18 controlled clinical trials and 19 before and after studies to assess the impact of  pharmacists reviewing and monitoring prescribing in outpatient clinics, primary care clinics or community pharmacies. Literature search to 1998.)



7.8j Further work is needed to evaluate whether the effectiveness of pharmacy discharge planning may be improved by providing information to general practitioners and community psychiatric nurses in addition to community pharmacists. One week post-discharge, both groups showed significant (p< 0.002) improvement in knowledge of medication from baseline and this improvement was maintained at 12 weeks. No significant difference was found between knowledge scores for the two groups on any occasion. Fewer medication problems were recorded for the intervention group. There was a trend for reduced readmissions for the intervention group, but this was not statistically significant (p = 0.065) Community pharmacists in receipt of plans were more likely to identify problems than other pharmacists.i

Caveat: The trial was conducted in a small sample.

i. Shaw H, Mackie CA, Sharkie I. Evaluation of effect of pharmacy discharge planning on medication problems experienced by discharged acute admission mental health patients. International Journal of Pharmacy Practice 2000; 8(2): 144-153
(Type II evidence – randomised controlled trial of 97 patients recruited from acute-admission psychiatric wards in Scotland to evaluate the effect of pharmacy discharge planning on the pharmaceutical care issues experienced by discharged mental health patients. Patients were allocated to either an intervention group (receiving a baseline pharmaceutical needs assessment, information about medicines and then a pharmacy discharge plan sent to their community pharmacy) or a control group.)


7.9 Risk-factors for non-compliance

7.9a Compared with nondepressed patients, the odds are 3 times greater that depressed patients will be noncompliant with medical treatment recommendations. Recommendations for future research include attention to causal inferences and exploration of mechanisms to explain the effects. Evidence of strong covariation of depression and medical noncompliance suggests the importance of recognising depression as a risk factor for poor outcomes among patients who might not be adhering to medical advice. The associations between anxiety and noncompliance were variable and their averages were small and nonsignificant. The relationship betweeen depression and noncompliance, however, was substantial and significant (OR 3.03, 95% CI 1.96-4.89).i

Caveat: Unpublished and non-English language were not sought.

i. DiMatteo MR, Lepper HS, Croghan TW. Depression is a risk factor for noncompliance with medical treatment: meta-analysis of the effects of anxiety and depression on patient adherence. Archives of Internal Medicine 2000; 160: 2101-2107
(Type IV evidence –systematic review and meta-analysis of 25 observational cohort studies (n=1466) correlating medical patients’ treatment noncompliance with their anxiety and depression. 10 studies considered medication adherence for patients with depression or schizophrenia. Literature review to 1998.)


7.9b Efforts to improve medication adherence in patients with schizophrenia should target relevant risk factors. Among the 10 reports that met a strict set of study inclusion criteria, we found a mean rate of non-adherence of 41.2%. 5 reports that met a stricter set of inclusion criteria had a mean non-adherence rate of 49.5%. In the 39 articles reviewed, factors most consistently associated with non-adherence included poor insight, negative attitude or subjective response toward medication, previous non-adherence, substance abuse, shorter illness duration, inadequate discharge planning or aftercare environment, and poorer therapeutic alliance. Findings regarding an association between adherence and medication type were inconclusive, although few studies explored this relationship. Other factors such as age, gender, ethnicity, marital status, education level, neurocognitive impairment, severity of psychotic symptoms, severity of medication side effects, higher antipsychotic dose, presence of mood symptoms, route of medication administration, and family involvement were not found to be consistent predictors of non-adherence.i

Caveat: Unpublished research was not sought.

i. Lacro JP, Dunn LB, Dolder CR, Leckband SG, Jeste DV. Prevalence of and risk factors for medication nonadherence in patients with schizophrenia: a comprehensive review of recent literature. Journal of Clinical Psychiatry 2002; 63(10): 892-909
(Type IV evidence – systematic review of 39 retrospective, cross-sectional, prospective and longitudinal studies. Literature search date unknown.)


7.9c Whilst selective serotonin reuptake inhibitors (SSRIs) do appear to show an advantage over tricyclic drugs in terms of total drop-outs, this advantage is relatively modest. This has implications for pharmaco-economic models, some of which may have overestimated the difference of drop-out rates between selective serotonin reuptake inhibitors and tricyclic antidepressants. These results are based on short-term randomised controlled trials, and may not generalise into clinical practice. SSRIs showed less participants dropping out compared to the tricyclic/heterocyclic group (OR 1.21, 95% CI 1.12 to 1.30). A statistically significant difference was found in total drop-outs between the selective serotonin reuptake inhibitors and the old tricyclics as well as the newer tricyclics. When SSRIs were compared to the heterocyclic antidepressants, there was a non significant difference favouring the selective serotonin reuptake inhibitors. The poor tolerability profile of the old tricyclics was explained by differences in drop-outs for side-effects, but not for inefficacy.i

i. Barbui C, Hotopf M, Freemantle N, et al. Treatment discontinuation with selective serotonin reuptake inhibitors (SSRIs) versus tricyclic antidepressants (TCAs). The Cochrane Database of Systematic Reviews 2000, Issue 4.
clsysrev/articles/CD002791/frame.html [accessed 29/07/05
(Type I evidence – systematic review and meta-analysis of 136 randomised double-blind trials to assess the comparative tolerability of selective serotonin reuptake inhibitors and tricyclic/heterocyclic antidepressant drugs. Literature search to 1999.)



7.9d This study supports recent meta-analyses of SSRIs versus tricyclic antidepressants in finding no significant differences in crude indices of compliance between fluoxetine and dothiepin, despite marked differences in side effect profile and dose regimen. In a secondary analysis using data from the Medication Event Monitoring System, both a survival analysis for length of time without a gap in medicine taking and a derived compliance index showed a significant advantage to fluoxetine. In both treatment arms patients with a superior compliance index were more likely to have improved in Hamilton depression scale scores by the last study visit.i

i. Thompson C, Peveler RC, Stephenson D, McKendrick J. Compliance with antidepressant medication in the treatment of major depressive disorder in primary care: a randomized comparison of fluoxetine and a tricyclic antidepressant. American Journal of Psychiatry 2000; 157(3): 338-343
(Type II evidence – randomised controlled trial of 152 patients with major depressive disorder, treated in 10 primary care clinics from the UK, allocated to receive either 20mg fluoxetine (mean age 37 years, 74% female) or 75mg dothiepin tablets (mean age 40 years, 67% female). 12-week follow-up.)


National Service Framework: key action 37

Raising the standard. Cardiff: Welsh Assembly Government, October 2005

Specific jointly agreed protocols are to be in place to ensure effective and seamless transitional arrangements for individuals (e.g. on transfer of care or discharge to the CMHT and the GP). [Key action 37 paragraph 27.2]

What information is available on referral to services to guide effective and seamless arrangements for individuals?

See Section 3.2 for preventing homelessness following discharge from hospital
See also Section 6.1 for communication between primary and secondary care and 6.7 for liaison psychiatry

The Statements The Evidence
7.10 Access to care and transitional arrangements
Gaps in the delivery of psychiatric services
7.10a Effective clinical bridging strategies can be used to avoid unnecessary gaps in the delivery of psychiatric services. Incorporating these strategies into routine care would enhance continuity of care, especially for some high-risk patients. Approximately two-thirds (65%) of the patients failed to attend scheduled or rescheduled initial outpatient mental health appointments after a hospital discharge. At high risk for unsuccessful linkage to outpatient care were patients with a persistent mental illness and those who had no prior public psychiatric hospitalisation, were admitted involuntarily, and had longer lengths of stay. Controlling for risk factors, three clinical interventions used during the hospital stay more than tripled the odds of successful linkage to outpatient care: communication about patients' discharge plans between inpatient staff and outpatient clinicians, patients' starting outpatient programs before discharge, and family involvement during the hospital stay.i

i. Boyer CA, McAlpine DD, Pottick KJ, Olfson M. Identifying risk factors and key strategies in linkage to outpatient psychiatric care. American Journal of Psychiatry 2000; 157: 1592-1598
(Type IV evidence – observational study analysing data from the medical records of 229 patients with a primary diagnoses of psychiatric disorder for whom inpatient staff had arranged referrals to outpatient programmes when discharged.)


7.10b This study suggests some important gaps in coverage by wider mental health services, which the current mental health agenda goes some way towards addressing. The majority (60.9%) of patients had had some specialist psychiatric contact in the follow-up period. Those with severe mental health problems formed a minority of presenters, but were heavy users of services. Coverage by Community Mental Health Team (CMHT) services was insufficient to prevent crisis in many cases. Improved access to broader community services is needed for those with life crises.i

i. Perry A, Hatfield B, Spurrell M. Specialist service use following psychiatric emergency presentation: an 18-month follow-up study. Health & Social Care in the Community 2002; 10(6): 457-463
(Type IV evidence – observational cohort study of referred psychiatric emergency cases in a 2-month time period. 189 referrals for 170 individuals were identified. After 18 months data were retrieved from hospital information systems on 169 individuals (mean age of 40 years, 53% male.).)


Referral to services
7.10c Guidelines are available with recommendations on a minimum essential dataset for communication from primary to secondary care. The recommended referral document is designed primarily for general practitioner referrals, but is intended to be suitable, with appropriate modification, for use by professions allied to medicine in any setting.i

i. Scottish Intercollegiate Guidelines Network. Report on a recommended referral document. SIGN Publication Number 31. Edinburgh: Scottish Intercollegiate Guidelines Network. 1998
(Type V evidence – guidelines based on the results of a national consensus conference and a review of the literature. The guideline is due for review. )


7.10d Referral to the Amalthea Project and subsequent contact with the voluntary sector results in clinically important benefits compared with usual general practitioner care in managing psychosocial problems, but at a higher cost. The Amalthea group showed significantly greater improvements in anxiety (average difference between groups after adjustment for baseline -1.9, 95% CI -3.0 to -0.7), other emotional feelings (average adjusted difference -0.5, 95% CI -0.8 to -0.2), ability to carry out everyday activities (-0.5, 95% CI -0.8 to -0.2), feelings about general health (-0.4, 95% CI -0.7 to -0.1), and quality of life (-0.5, 95% CI -0.9 to -0.1). No difference was detected in depression or perceived social support. The mean cost was significantly greater in the Amalthea arm than the general practitioner care arm (£153 versus £133, p=0.025).i

i. Grant C. A randomised controlled trial and economic evaluation of a referrals facilitator between primary care and the voluntary sector. British Medical Journal 2000; 320(7232): 419-23
reprint/320/7232/419  [accessed 29/07/05]

(Type II – randomised controlled trial and economic evaluation of 161 patients identified by their general practitioner as having psychosocial problems, from 26 general practices in Avon. 90 patients (mean age 40.8 years, 72% female) were allocated to the Amalthea Project (a liaison organisation facilitating contact between voluntary organisations and patients in primary care) and 71 patients (mean age 45.5 years, 79% female) received routine general practitioner care. 4-months follow-up.)


7.10e Referrers want forensic assessments to be of a high quality and to be performed quickly. Assessments and forensic reports were completed (and questionnaires sent to referrers) in 63% of total referrals (32 out of 51). The response rate to the questionnaire was 81% (26 out of 32). Many referrers wanted the assessments and report to be completed in 2 weeks. Most referrers were satisfied with the quality of the report received and the majority were happy with the risk assessment.i

i. Papanastassiou M, Roche S, Boyle J, Baxter R, Chesterman P. A survey of referrers' satisfaction with a regional forensic psychiatric service: what do they want? Psychiatric Bulletin 2003; 27: 96-98
(Type IV evidence – survey study of all referrals to 2 teams at the North West Thames Regional Secure Unit over a 6-month period. Self-report questionnaires were completed by 26 referrers, in cases where an assessment and forensic report had been completed.)


7.10f Among patients with psychosis, having a diagnosis of schizophrenia and being male increases the likelihood of special hospital admission. Suggestions that ethnic minority patients are much more likely to have engaged in serious violence and need high-security placement were not borne out. Schizophrenia was the almost invariable diagnosis for all special hospital patients. White patients in the community sample were significantly more likely to have affective components to their illness compared with African-Caribbean patients; unlike those in special hospitals. There was a small excess in the proportion of African-Caribbean patients in the special hospital group, controlling for diagnosis, gender and locality. Men were overrepresented in this groupi.

i. Walsh E, Leese M, Taylor P et al. Psychosis in high-security and general psychiatric services: report from the UK700 and special hospitals' treatment resistant schizophrenia groups. British Journal of Psychiatry 2002; 180: 351-357
(Type IV evidence – observational study comparing a national sample of 905 high-security hospital residents (mean age 38 years, 88.9% male) with a community sample from the UK700 trial of 708 patients in contact with general psychiatric services (mean age 36 years, 57% male). Sociodemographic and diagnostic measures were gathered for the special hospital sample from case records and notes, and from case notes and patient interviews for the UK700 sample.)


7.10g It may be more realistic to plan future services on the basis that only 9% of patients are misplaced in special hospitals, rather than previous estimates that appear to have guided current policy. Patients detained under the legal category of psychopathic disorder present particular problems and there is a need to develop appropriate facilites at medium secure level. In the meantime, no patients should be admitted to high security without consultation with the catchment area service and a jointly agreed plan for future rehabilitation. i

See also Section 6.8 for levels of secure care

i. Sayal K, Maden A. The treatment and security needs of patients in special hospitals: views of referring and accepting teams. Criminal Behaviour and Mental Health 2002; 12(4): 244-253
(Type IV evidence – study comparing the views of the referring and receiving team on the security needs of all special hospitals patients (n=119) from two London health authorities. Views of both teams were collected from patient case notes and treatment needs for 80 patients with schizophrenia  was supplemented with mental state examinations.)



7.10h The results show that professionals and non-health professionals, such as the patient themselves, carers or a voluntary group, are equally likely to make inappropriate referrals, but, overall, open access was used efficiently by both groups. Analysis of emergency assessments showed that 46% of psychiatric patients were referred directly to the service from a non-health professional source. A total of 54% were referred from a health professional. The results also indicated that open access to the service enabled early intervention for high-risk client groups. 50% of emergency referrals required medical intervention and 45% of this group had been referred by a non-professional source.i

Caveat: Patient characteristics have not been reported.

i. Smith C, Embling S, Price P. Unrestricted access to mental health services. Nursing Standards 2002; 16(52): 33-36
(Type IV evidence – audit study of 1544 mental health assessments conducted at a nurse-led, community-based, acute psychiatric assessment and treatment centre in England, collated over 12 months. Only those referrals assessed as emergencies (n=446) were analysed. Comparisons were made for emergency referrals from professionals and non-health professionals in terms of appropriateness. Analysis of data focused on risk, diagnosis and outcome of each referral.)


7.10i Results of an audit of a service for children with learning disabilities, within a child and adolescent mental health team (CAMHS), suggests that the aims of the service are being achieved but given the fact that the numbers of new referrals significantly outweigh the discharge rate, it is concluded that a greater emphasis is placed on liaison, consultation and joint working with other agencies. Overall, 43 different referrers were identified. These included GPs, paedatiricians, social workers and educational psychologists. The majority of referrals were made by paediatricians (45%). A wide range of problems were reported in the referral letters. 32 different requests were made with regard to input. The most frequent requests were for help with general behaviour problems (20%), and autism assessments (24%). Just over half of the children attended special schools (58%). Only 6 children and their families failed to attend for their initial appointment. Waiting-lists were reduced from one year to approximately one month.i

Caveat: Methods of data collection and analysis have not been reported.

i. Green K, Williams C, Wright B, Partridge I. Developing a child and adolescent mental health service for children with learning disabilities. Psychiatric Bulletin 2001; 25(7): 264-267
(Type IV evidence – longitudinal audit study of 111 children (mean age 9 years, 71% male) attending a service for children with learning disabilities within a child and adolescent mental health team in York. Data were collected between 1990 to 1999.)


7.10j Providers from the mental health and child welfare sectors have more professional training in mental health and are more likely to receive inservice training. Inservice training should be offered to all who work with youths. Structural equation models demonstrate that provider assessment of youths' mental health problems is the largest and provider knowledge of service resources the second largest determinant of service provision. Youths' self-reported mental health is not positively associated with increased services and is only minimally associated with provider assessment of their problems. Training (both professional and inservice) contributes to higher assessments of youths' problems and greater resource knowledge, which is associated with increased service provision.i

Caveat: The response rate to the provider’s survey was 61%. The results of this study may not be generalisable to a UK setting.

i. Stiffman AR, Hadley-Ives E, Dore P, et al. Youths' access to mental health services: the role of providers' training, resource connectivity, and assessment of need. Mental Health Services Research 2000; 2(3): 141-154
(Type IV evidence – combined data analysis with follow-up interviews. 792 adolescents (mean age 15.3 years, 57% females), involved with public health, juvenile justice, child welfare or education service sectors in America were interviewed in 1994 and 1996. 364 of 533 providers, identified by 282 youths in 1996, received mailed questionnaires, concerning service need, service use, and provider knowledge and behaviour. Questionnaires were followed-up by telephone interviews.)


7.10k The general practitioners made an average of 7.1 community referrals and 3.8 of their patients were hospital referrals. There was a large variation in the number of referrals (range 1-45). However, there was a significant positive relationship between the number of hospital and community referrals for each GP (p=0.001). These findings support the hypothesis that there would be a positive correlation between hospital and community referrals, and suggest that it may be useful to use the ratio, hospital to community referrals, to identify GPs with abnormal referral patterns.i

i. Butler R, Oyewole D, Pitt B. What is the relationship between general practitioners' community referrals, and hospital referrals to an old age psychiatric service? Aging & Mental Health 2000; 4(1): 79-81
(Type IV evidence – data analysis of 674 first referrals (mean age 78.6 years, 62% female), from 62 general practitioners, to an old age academic, hospital-based psychiatric service in London between 1989 and 1995.)



7.10l Variation in referral rates remains largely unexplained. Targeting high or low referrers through clinical guidelines may not be the issue. Rather, activity should concentrate on increasing the number of appropriate referrals, regardless of the referral rate. Pressure on GPs to review their referral behaviour through the use of guidelines may reduce their willingness to tolerate uncertainty and manage problems in primary care, resulting in an increase in referrals to secondary care. The use of referral rates to stimulate dialogue and joint working between primary and secondary care may be more appropriate. Patient characteristics explain <40% of the observed variation; practice and GP characteristics <10%. The availability of specialist care does affect referral rates, but its influence on the observed variation of referral rates is not known. Intrinsic psychological variables are important. GPs who are less tolerant of uncertainty or who perceive serious disease to be a more frequent event may refer more patients. There is a lack of consensus about what constitutes an appropriate referral, and the use of guidelines has had only limited success in altering referral behaviour.i

i. O'Donnell CA. Variation in GP referral rates: what can we learn from the literature? Family Practice 2000; 17(6): 462-471
(Type V evidence – expert opinion.)


National Service Framework: key action 37

Raising the standard. Cardiff: Welsh Assembly Government, October 2005

Communication within and between services must be robust. There are to be effective protocols in place for communication of risk and sharing information both to the individual and to others including those providing services. [Key action 35 paragraph 31.1]

Performance Target: By March 2007 LAs/LHBs/NHS Trusts to develop local protocols for communication of risk and transfer of care, the sharing of care within and between agencies and to ensure that people achieve equality of access to the range of services. Protocols should cover the following range of services:

  • Young people in transition to adult services

  • People with a co-occuring learning disability
  • Older people in transition from adult services
  • People in contact with criminal justice services
  • People with co-occuring mental health and substance misuse

How best can communication withina nd between services be established and maintained?
What are the needs of people with a co-occuring learning disability and how can they be managed? 

See also Chapter 7 for the following:

  • Communication / liaison with child and adolescent mental health services – Section 7.17
  • Mental health care and criminal justice services – Section 7.14-7.16
  • Rish management – Section 7.3
The Statements The Evidence
7.11 Learning disabilities and mental health problems
Needs of people with mental health problems and intellectual, developmental or learning disabilities

7.11a The Camberwell Assessment of Need for Adults with Developmental and Intellectual Disabilities (CANDID) is a brief, valid and reliable needs assessment instrument for adults with learning disabilities and mental health problems. CANDID scores were related with both Disability Asssessment Schedule (p<0.05) and the Global Assessment of Functioning (p<0.01). Correlation coefficients for interrater reliability were 0.93 (user), 0.90 (carer) and 0.97 (staff ratings); for test-retest reliability they were 0.71, 0.69 and 0.86 respectively. Mean interview duration was less than 30 minutes.i

Caveat: The sample for the reliability study  was relatively small (40 service users) and from only two sites.

i. Xenitidis K, Thornicroft G, Leese M et al. Reliability and validity of the CANDID – a needs assessment instrument for adults with learning disabilities and mental health problems. British Journal of Psychiatry 2000; 176: 473-478
(Type IV evidence – study to develop and assess validity of the Camberwell Assessment of Need for Adults with Developmental and Intellectual Disabilities (CANDID). Concurrent validity was tested using the Global Assessment of Functioning and the Disability Assessment Schedule. Test – retest and interater reliability were investigated using 40 adults with learning disabilities and mental health problems in an outer London borough.)


7.11b A high percentage of needs were identified in areas such as communication and language, socially embarrasing behaviour, domestic skills, money management, accommodation and social life. In the clinical domain, needs were identified in all areas except psychotic symptoms, sensory impairment, drug and alcohol abuse and side effects of medication. In the social domain, communication skills and language, domestic skills, money management, hygiene and dressing, accommodation, employment and social life top the list in terms of the number of needs identified. The cardinal problem was persisting despite intervention in four areas, which include domestic skills (4 people), money management (2 people) and for 1 person each in the area of communication skills and language and mobility and use of amenities.i

i. Ragharan R. An investigation into the needs of people with learning disabilities and mental health disorders (dual diagnosis). Thesis.Oxford: Oxford Brookes University. 2000
(Type IV evidence – qualitative study to identify and adapt an appropriate methodology for systematically assessing the needs of people with dual diagnosis, and to identify the needs of people with a dual diagnoisis. 35 individuals from England (mean age 41 years, 51% female) completed a learning disability version of the Cardinal Needs Schedule (LCDNS).



7.11c If targets for reducing ill health caused by mental illness are to be met, further prospective, phenomenological research into psychiatric disorders is required for people with intellectual disability (ID) who also have sensory impairments (SIs). Extensive educational programmes also need to be encouraged. In 12 patients (75% of those referred), congenital rubella accounted for their SIs. The majority of patients (n=10) were referred because of self-injurious behaviour (SIB) and aggression. In 10 patients accurate and reliable ICD-10 diagnoses could not be made because of their ususual behavioural presentation and degree of ID. In several of these cases, depression was the postulated diagnosis. 9 cases were treated with antidepressants, 5 underwent environmental changes and 2 had medication reduced. All showed some improvement.i

Caveat: The number of people followed was very small.

i. Carvill S, Marston G. People with intellectual disability, sensory impairments and behaviour disorder: A case series. Journal of Intellectual Disability Research 2002; 46: 264-72
(Type IV evidence – observational case series in Birmingham of 18 patients (aged 27-34 years, 67% male) who were clinically assessed, and where specific diagnoses could be made, appropriate psychopharmacological interventions were employed. Patients’ progress was monitored via monthly psychiatric visits.)


7.11d People with learning disabilities are vulnerable to the same mental health problems as the general population. However services for them are less than adequate. Mental health services are often reluctant to accept them, while learning disability services tend to lack the skills and resources needed to meet their additional mental health needs. Inadequacy of diagnostic and assessment processes too may hamper the provision of appropriate treatment.i

i. Coyle D. Meeting the needs of people with learning disabilities and mental health problems: a review. Mental Health Care 2000; 3(12): 408-411
(Type V evidence – expert opinion.)


Management of people with mental health problems and intellectual, developmental or learning disabilites
7.11e Reviewers found no randomised controlled trial evidence to guide the use of antipsychotic medication for those with both learning disability and schizophrenia. Until the urgent need for randomised controlled trials is met, clinical practice will continue to be guided by extrapolation of evidence from randomised controlled trials involving people with schizophrenia but without learning disability and non-randomised trials of those with learning disability and schizophrenia. The one relevant randomised trial identified by the searches included 4 people with a dual diagnosis of schizophrenia and learning disability, but results were available for only 2. The groups to which the other 2 people were allocated were unclear. In order to display the data, too many assumptions would have to have been made about these other 2 people and any results would be uninformative and potentially misleading.i

i. Duggan L, Brylewski J. Antipsychotic medication versus placebo for people with both schizophrenia and learning disability. The Cochrane Database of Systematic Reviews 2004, Issue 4.
clsysrev/articles/CD000030/frame.html [accessed 29/07/05
(Type I evidence – systematic review of 1 double-blinded randomised controlled trial. Outcomes included: leaving the study early, clinical ratings (non-standardised scale), side-effects, attendance at occupational therapy and physiological measures. Literature search to 1999.)


7.11f Participants felt that a consultancy service (for example, being able to phone to ask for advice rather than making a referral) would be a useful addition to the individual referral system and make better use of the psychologist’s expertise and time. Participants also requested that more time be spent training staff to deal effectively with problems themselves. There also seemed to be a need to educate people about what psychologists are able to provide in this region. There was some disparity between what group members asked for in the focus groups and what they actually refer for.i

Caveat: The sampling method and methods of data collection and analysis are not reported. Participant characteristics have also not been reported.

i. Waddell H, Evers C. Psychological services for people with learning disabilities living in the community: focus group views. Clinical Psychology Forum 2000; 141: 34-38
(Type IV evidence –
UK qualitative study of 7 focus groups to explore professionals’ and carers’ experience and perceptions of a psychology service for people with learning disabilities living in the community. Focus groups lasted 30-60 minutes and had between 2 to 10 participants in each. Participants included two groups of people with learning disabilities, two groups of social education centres staff, community home managers, parents and carers, and a team of social workers and a community nurse. Discussions were tape recorded. GPs (n=29) in five practices were sent a postal questionnaire.)



7.11g Using the DSM-IV criteria, which require at least 5 out of 9 symptoms, helps to specify Borderline Personality Disorder (BPD) and prevent overdiagnosis. 3 symptoms – impulsivity, affective instability and inappropriate intense anger or difficulty controlling anger – are comonly seen in individuals with Developmental Disability (DD), frequently in a cluster, but are clearly not sufficient for the diagnosis of BPD. A fourth symptom, self-mutilating behaviour, is also commonly seen in individuals with DD, often as a generalised response to stress and often as part of an autistic disorder. 4 out of 9 symptoms at most, might be present in many individuals with DD; however, they still would not meet the DSM-IV criteria.

Pharmacological treatment must be custom tailored to each individual person with BPD based on his or her presenting symptoms. Furthermore, pharmacological treatment alone is generally not sufficient but must be combined with psychotherapy or behavioural strategies in individuals with or without (DD). An integrated team approach is essential to avoid splitting, provide consistency and to educate members of the team who may have limited experience with BPD.i

i. Mavromatis M. The diagnosis and treatment of Borderline Personality Disorder in persons with developmental disability: three case reports. Mental Health Aspects of Developmental Disabilities 2000; 3(3): 89-97
(Type IV evidence – case series report to summarise the diagnostic criteria for Borderline Personality Disorder (BPD) and its application to individuals with Developmental Disability (DD). 3 individual case reports are described to illustrate the range and complexity of both diagnostic and treatment issues.)


7.11h The method of diagnosis for people with severe and profound intellectual disability (ID) remain debatable, with some authors advocating adherence to standard criteria, others suggesting adding criteria to the standard ones and yet others believing that substitute criteria are called for. However, for those with mild to moderate ID, a consensus is emerging that standard diagnositc criteria are appropriate. There has been progress in examining some of the symptoms which might constitute depression in people with ID. New diagnostic criteria issues by the Royal College of Psychiatrists are to be welcomesd. Although new rating scales have emerged, there is as yet no gold standard disagnostic tool for depression amongst peole with ID.i

i. McBrien JA. Assessment and diagnosis of depression in people with intellectual disability. Journal of Intellectual Disability Research 2003; 47(Part 1): 1-13
(Type V evidence – expert opinion.)


7.11i De-institutionalisation has led to the expectation that more complex and challenging people be placed in the community. This study suggests the community to be, as yet, unready to cope with the needs of more complex and challenging people. Three quarters of the patients (n=181) were found not to be ready for discharge, the remaining 66 cases were, however, ready for discharge to appropriate placements. 22 of those ready for discharge had a discharge planned, but 44 experienced delays and a ‘lack of resources’ was common to all of this group.

Delay was ascribed to a lack of suitable accommodation (n=34), insufficient funding (n=10), carers who were deemed unable to cope (n=17), insufficient clinical support (n=11) and a lack of suitable educational placement (n=13). At 16-months follow-up, all of those delayed by the last 2 factors, as well as 70% of those with insufficient funding, had been discharged. However, only 39% (n=13) of those who had been delayed by a lack of accommodation had achieved discharge, suggesting that this was less easily resolved.i

i. Watts RV, Richold P, Berney TP. Delay in the discharge of psychiatric in-patients with learning disabilities. Psychiatric Bulletin 2000; 24: 179-181
(Type IV evidence – follow-up census study of psychiatric in-patients (mean age 34 years) from a learning disability service in England, to identify factors that delay discharge. A census was completed, categorising in-patients as unready for discharge (n=181); discharge planned within 12 weeks (n=22); or ready for discharge but experiencing delay (n=44).The latter were followed-up 16 months later.)



7.11j Comprehensive multidisciplinary assessment and treatment are particularly important for clients who present with learning disabilities and mental health problems. This should not be restricted to professionals who work in services for people with learning disabilities and should include a wider network of professionals from statutory services and the independent sector. This arrangement can be complex as different agencies and even different departments of the same agency may differ fundamentally in their philosophy, vision and understanding of their own and other’s roles and responsibilities. It is vital that these differences are recognised and dealt with so that they do not become a barrier to accessing best practice.i

i. Martin P. Learning disabilities and mental ill health: care plans. Nursing Times 2001; 97(29): 42-43
(Type V evidence – expert opinion.)


7.12 Substance misuse and mental health problems
See also Section 6.12

The Statements The Evidence
7.12a The problems posed by substance misuse in the context of severe mental illness will not go away. The current momentum for integrated programmes is not based on good evidence. Implementation of new specialist substance misuse services for those with serious mental illnesses should be within the context of simple, well designed controlled clinical trials. 6 relevant studies, 4 of which were small, were identified. In general, the quality of design and reporting was not high. Clinically important outcomes such as relapse of severe mental illness, violence to others, patient or carer satisfaction, social functioning and employment were not reported. There is no clear evidence supporting an advantage of any type of substance misuse programme for those with serious mental illness over the value of standard care.i

i. Jeffery DP, Ley A, McLaren S, Siegfried N. Psychosocial treatment programmes for people with both severe mental illness and substance misuse. The Cochrane Database of Systematic Reviews 2000, Issue 2.
clsysrev/articles/CD001088/frame.html [accessed 29/07/05
(Type I evidence – systematic review of 6 randomised controlled trials. Literature search to 1998.)


7.12b Alcohol-abusing patients taking clozapine experienced significant reductions in severity of alcohol abuse and days of alcohol use while on clozapine. For example, they averaged 54.1 drinking days during 6-month intervals while off clozapine and 12.5 drinking days while on clozapine. They also improved more than patients who did not receive clozapine. At the end of the study, 79.0 % of the patients on clozapine were in remission from alcohol use disorder for 6 months or longer, while only 33.7% of those not taking clozapine were remitted. Findings related to other drugs in relation to clozapine were also positive but less clear because of the small number of patients with drug use disorders. The use of clozapine by patients with co-occurring substance disorders deserves further study in randomised clinical trials.i

Caveat: This study was limited by the lack of prospective, standardised measures of clozapine use. It is unclear whether an intention-to-treat analysis was used.

i. Drake RE, Xie H, McHugo GJ, Green AI. The effects of clozapine on alcohol and drug use disorders among patients with schizophrenia. Schizophrenia Bulletin 2000; 26(2): 441-9
(Type II evidence – retrospective analysis of 151 patients in America (mean age 32.4 years, 77.5% male) with schizophrenia and co-occurring substance use disorder assigned to receive dual disorder services through either assertive community treatment or through standard case managemement. All participants were assessed prospectively every 6 months over 3 years for psychiatric symptoms and substance use.)


7.12c Although motivational interviewing appears feasible among in-patients in psychiatric hospital with co-morbid substance use disorders, more extensive interventions are recommended, continuing on an out-patient basis, particularly for cannabis use. There was a modest short-term effect of the motivational interview on an aggregate index of alcohol and other drug use (polydrug use on the OTI) compared to the control group (F1,105= 4.47, p=0.04). Cannabis use remained high among the sample over the 12-month follow-up period (58.1%).i

Caveat: It is unclear whether an intention-to-treat analysis was conducted. The follow-up rate was low (71.9%).

i. Baker A, Lewin T, Reichler H, et al. Evaluation of a motivational interview for substance use within psychiatric in-patient services. Addiction 2002; 97(10): 1329-1337
(Type II evidence – randomised controlled trial of 160 psychiatric in-patients in Australia (mean age 30.8 years, 75% male) with coexisting alcohol and other drug (AOD) problems assigned to receive either an individual motivational interview or a self-help booklet (control condition). 12-month follow-up.)


7.12d Data suggests that social support can be improved through alcohol treatment in individuals with coexisting social phobia, particularly if social workers emphasise skills training and facilitate involvement in 12-step groups. For men, there was significant improvement on 2 measures of social support regardless of treatment group (Social Support-Friends scores, F1,251=24.70, p<0.001). Post-treatment scores (m=4.03) were higher than baseline scores (m=3.56). The effect size was large (η2=0.91). Women who received Cognitive-Behavioural Therapy (CBT) or Twelve Step Facilitation (TSF) had better support outcomes than women who received Motivational Enhancement Therapy (MET) (t(95)=1.42, p=0.02).i

i. Thevos AK, Thomas SE, Randall CL. Social support in alcohol dependence and social phobia: treatment comparisons. Research on Social Work Practice 2001; 11(4): 458-472
(Type II evidence – retrospective secondary analysis of 397 participants (mean age 39 years; 73% male) seeking treatment for alcoholism and who participated in Project MATCH, a multisite clinical trial for alchoholism conducted in America. Participants were allocated to receive either CBT, TS, or MET over 12- weeks. 15-month follow-up.)



7.12e The findings suggest that direct family support may help people with dual disorders to reduce or eliminate their substance use. Family economic support was associated with substance abuse recovery in bivarite and regression analyses. Caregiving hours were significanty associated with changes in psychiatric symptoms. Regression results showed that higher average family expenses on behalf of the study participant were associated with reductions in substance use (average monthly family expenditures 0.318 p=0.003) and more caregiving hours were significantly associated with substance use (average monthly caregiving hours 0.006 p=0.03). Further research is needed to confirm this connection and to establish the mechonisms by which support is useful.i

Caveat: Although the original study had 203 participants, the data presented in this paper is for 151 patients due to missing data for family caregiving and/or psychiatric symptoms.

i. Clark RE. Family support and substance use outcomes for persons with mental illness and substance use disorders. Schizophrenia Bulletin 2001; 27: 93-101
(Type II evidence – secondary analysis of data from a 3-year randomised controlled trial of 203 patients (mean age 34 years, 74% male) in treatment for dual disorders in
America. Caregivers for 174 participants completed structured interviews at 6-month intervals on type and amount of care and economic support provided.)


7.12f A cognitive-behavioural treatment (CBT) programme for panic and agoraphobia, in addition to a regular alcoholism treatment programme, had not been more effective than the regular alcohol treatment programme in reducing problem drinking in those with panic disorder. Abstinence from drinking, and anxiety and mood symptoms improved after treatment in all of the groups; there were few differences in outcome between the groups.i

i. Bowen RC, D'Arcy C, Keegan D, Senthilselvan A. A controlled trial of cognitive behavioural treatment of panic in alcoholic inpatients with comorbid panic disorder. Addictive Behaviors 2000; 25(4): 593-597
(Type II evidence – randomised controlled trial of 231 patients (mean age 32.8 years, 64% male) in an alcohol treatment centre in Canada, with anxiety and phobic symptoms. 146 patients received CBT intervention in addition to the regular alcoholism treatment, and 85 patients received the regular programme. 12-months follow-up.)

7.12g Routine urine drug screening in a psychiatric emergency service did not affect disposition or the subsequent length of inpatient stays. The results do not support routine use of drug screens in this setting. As for accuracy of physicians' suspicion of substance use, positive drug screens were recorded for 10.2% of the 198 patients in the mandatory-screen group who did not admit drug use or for whom physicians did not expect drug use. A total of 39.3% of the patients who were suspected of use and 88.2% of those who admitted use had positive drug screens. Only 20.8% of patients who denied substance use had positive screens.i

i. Schiller MJ, Shumway M, Batki SL. Utility of routine drug screening in a psychiatric emergency setting. Psychiatric Services 2000; 51(4): 474-478.
cgi/reprint/51/4/474 [accessed 29/07/05
(Type II evidence – randomised controlled trial of 392 patients presenting to an urban psychiatric emergency service in America assigned to either a mandatory-screen group (64.7% male) or a usual-care group (75.8% male). Physicians ordered screens based on clinical judgment. Additional screens were performed without physicians' knowledge for patients in the mandatory-screen group for whom no screen was ordered.)


7.12h The negative outcomes for mentally ill adults with substance-dependent subjects in both programmes suggest that the Intensive Clinical Case Management (ICCM) and and Expanded Brokerage Case Management (EBCM) models were relatively ineffective for these patients. Substance dependence predicted negative outcomes independent of the case management intervention. ICCM was the superior treatment for subjects who were not dependent on substances. Fewer of them required psychiatric hospitalisation in the 6-month post discharge period than in the 6-month period before hospital admission.i

Caveat: The results of this study may not be generalisable to a UK setting. 70.1% follow-up.

i. Havassy BE, Shopshire MS, Quigley LA. Effects of substance dependence on outcomes of patients in a randomized trial of 2 case management models. Psychiatric Services 2000; 51(5): 639-644.
cgi/reprint/51/5/639 [accessed 29/07/05
(Type II evidence – randomised controlled trial of 268 seriously mentally ill adults with and without substance dependence admitted to the psychiatric service of San Francisco General Hospital (age 18-59 years, 65.3% male). Patients were stratified by substance dependence status and allocated to either Intensive Clinical Case Management (ICCM) or Expanded Brokerage Case Management (EBCM). 6 months follow-up.)



7. 13 Communication/ liaison with mental health services for older people

7.13a This research suggests that psychosocial interventions that involve collaboration between carers and residents, supported by a Community Mental Health Team, may have an important part to play in supplementing medical management of depression in residential care homes. The training programme was positively evaluated by the recipients, the trainers and the researcher who observed it. The ability of care staff to detect depression improved significantly over time, and depression was reduced to below case-level in 7 of the 8 depressed residents who participated in the care-planning intervention.i

Caveat: Carers were renumerated from their employer’s training budget. The results are based on a small sample.

i. Moxon S, Lyne K, Sinclair I. Mental health in residential homes: a role for care staff. Ageing & Society 2001; 21(1): 71-93
(Type III evidence – 2 linked studies assessing the feasibility of providing mental health training (delivered by a community mental health team for the elderly) for care staff in 2 UK residential homes for the elderly. In study 1, 20 care staff completed training to detect depression in residents, which was evaluated in 36 residents (mean age 85.7 years, 82.1% female). In study 2, the same 20 care staff were trained to implement a care-planning intervention for the 10 residents with depression identified from the previous study.)


7.13b A coordinated management and educational initiative resulted in marked improvement in basic medical and psychiatric assessment and more integrated care. These changes did not require expansion of specialist services. There were significant improvements in mental health staff communication with general practitioners (73% prior to the improvement programme and 97% after implementation). There were also significant improvements in medical (43% to 92%), neurocognitive (37% to 84%) and behavioural (e.g. suicidal ideation: 78% to 100%) assessments. The most change occurred in the adult community-based treatment services.i

Caveat: The management of only a small number of patients are assessed, of which 91% were female.

i. Mutch C, Tobin M, Hickie I. Improving community-based services for older people with depression: the benefits of an educational and service initiative. Australia & New Zealand Journal of Psychiatry 2001; 35(4): 449-454
(Type III evidence – before and after study implementing a quality improvement strategy within an Australian district mental health service (consisting of psychogeriatric and inpatient services and community care). Mental health staff received education about evidence-based treatment of depression and were provided with standardised assessment and treatment tools. Following implementation, the management of 44 clients with major depression (mean age 65.4 years; 91% female) were assessed during 1999, and compared with  99 subjects (mean age 68.9 years, 69% female) prior to implementation in 1995.)

7.13c GPs strongly reported the need for adequate, long term care, together with support for both family carers and healthcare workers, the importance of a multi-disciplinary approach, better liaison and communication with Social Services and the need to take advantage of the opportunity for planning at several different levels. Other needs were rapid access to care, clear diagnosis, special needs of young patients with Alzheimer's disease, joint guidelines, and an increased role for Community Psychiatric Nurses (CPNs). GPs valued the role of voluntary organisations and help given to them by Consultant Psycho-geriatricians.i

Caveat: Only 118 GPs replied to the survey (62.7%).

i. Williams I. What help do GPs want from specialist services in managing patients with dementia? International Journal of Geriatric Psychiatry 2000; 15: 758-761
(Type IV evidence – questionnaire survey. 188 GPs from England were invited to complete a 2-part questionnaire in relation to their needs from a specialist psycho-geriatric service.)


7.13d In old age mental health services over half of the respondents reported joint screening arrangements for health and social care, almost four-fifths reported both joint criteria for the allocation of key workers and a clear definition of monitoring responsibilities. Of the latter over two-fifths were reported as being the same in care management and the care programme approach (CPA). 46% of respondents provided a specialist service for people with dementia. Three-fifths of respondents reported that they did not apply CPA to people with dementia who were in receipt of care management or did so in less than 20% of cases. Where the CPA was applied it was more likely that a priority would be accorded to care management. A quarter of respondents reported the shared use of assessment documentation for people with dementia.i

Caveat: Characteristics of questionnaire respondents have not been reported.

i. Hughes J, Stewart K, Challis D, Darton R, Weiner K. Care management and the care programme approach: towards integration in old age mental health services. International Journal of Geriatric Psychiatry 2001; 16(3): 266-272
(Type IV evidence – postal questionnaire survey of 131 local authorities in England to examine the relationship between care management arrangements and the Care Programme Approach (CPA). Follow-up questionnaires were sent to 85% of respondents relating to older people’s services and to services for those with mental health problems.)



National Service Framework: key action 38

Raising the standard. Cardiff: Welsh Assembly Government, October 2005

There are tp be arrangements in place to support criminal justice services including prisons and youth offending teams. Other provision is to include diversion from custody and in-reach into prisons to ensure as seamless care as possible for offenders with mental health problems. There should be clear protocols to manage individuals who have a history of offending. [Key action 38 paragraph 32.1]

What arrangements can be made to support criminal justice services in managing offenders with mental health problems?
What evidence is available to inform the care of offenders with mental health problems?

The Statements The Evidence
7.14 Supporting and collaborating with criminal justice services
7.14a An inter-agency model where primary mental health workers (PMHW) work within a Youth Offending Team (YOT) may be a useful way of strengthening the links between specialist Child and Adolescent Mental Health Services (CAMHS) and YOTs, and may provide an accessible, responsive and effective service to a needy group of young people. In addition to the anticipated concerns about oppositional/aggressive behaviour, young people were referred for a range of mental health problems. There were high levels of emotional problems (72.5%), self-harm (40%), peer (45%) and family relationship difficulties (90%), and school non-attendance (62.5%). PMHWs offered a range of direct interventions, as well as consultation to YOT staff.i

i. Callaghan J, Pace F, Young B, Vostanis P. Primary mental health workers within youth offending teams: a new service model. Journal of Adolescence. 2003; 26(2): 185-99
(Type IV evidence –
UK survey study of 40 (mean age 15.5 years; 82.5% male) young people consecutively referred to PMHWs during the initial 6-month phase of PMHWs being in post. Interview checklists were completed for all participants and YOT staff completed satisfaction questionnaires after seeing 20 young people (mean age 15 years; 80% male) for consultation.)


7.14b Mental health service provision through primary mental health workers is a useful model for interagency partnerships for high-risk client groups with multiple and complex mental health needs. 4 themes were identified: previous experiences of specialist mental health services; issues of interagency working; the role of the primary mental health worker within the YOT; and recommendations for the future. Overall, the clinical component of the role (assessment and intervention), and the accessibility and responsiveness of the mental health staff were consistently valued, while there were mixed responses on role definitions within the team, consultation and training.i

i. Callaghan J, Young B, Pace F, Vostanis P. Mental health support for youth offending teams: a qualitative study. Health & Social Care in the Community 2003; 11(1): 55-63
(Type IV evidence – qualitative study of 4 focus groups with 17 youth offending team (YOT) staff members in England. Data examing the perceptions of YOT professionals about the role of primary mental health workers within their team were analysed according to the constant comparative method.)


7.14c Learning to work collaboratively as part of a professional team is essential to enhance multidisciplinary teamwork. There is little in professional training curricula that addresses this area of practice. It cannot be assumed that multidisciplinary collaboration will occur without managerial support, interprofessional learning and members’ willingness to engage in and monitor the process. Personal role clarity scores ranged from 7 to 35. The mean of the medical group (28.2) was significantly higher than that of the support group (24.2), (difference=4.6, p<0.05). Staff from high security sites (mean=27.1) had greater team role clarity than staff from low secure sites (mean 23.9) (difference=3.2, range 7 to 35, p<0.05). Medical staff (mean 28.2) had greater team role clarity than therapy staff (mean 24.0) (difference=4.2, p<0.05). Staff from high secure sites (mean 35.7) indicated significantly greater team identity than staff from low secure sites (mean=32.7) (difference=3.1, range 8 to 40). Medical staff (mean=37.7) and nursing staff (mean=35.0) both indicated greater team identity than support staff (mean 30.9) (difference between medical and suport staff=6.8) (difference between nursing and support staff=4.1). Medical staff also had significantly greater team identity than therapy staff (mean=32.5) (difference=5.2).

Five broad themes emerged from the team interviews: teams provide a number of functions for professional members; teams value client engagement; teams from each of the different security levels have different needs; teams recognise their own knowledge and skill deficits; teams value learning about being a team.i

i. Whyte L, Brooker C. Working with a multidisciplinary team in secure psychiatric environments. Journal of Psychosocial Nursing & Mental Health Services 2001; 39(9): 26-34
(Type IV evidence – quantitative and qualitative study. 233 staff from 20 multidisciplinary teams in the UK (prison health care n=4, low secure sites n=6, medium secure sites n=5, high secure sites n=5) completed 29-item questionnaires and 4 teams from each level of secure care were interviewed in focus groups. Teams were chosen based on questionnaire scores for the highest levels of clarity related to the five aspects of team functioning.)



7.14d Improved liaison between sector psychiatrists and local police may be of value in the earlier identification and treatment of the mentally ill. It was widely accepted by police officers that mental illness occurred commonly, can be effectively treated in the community and that the main risk of harm is to patients themselves. There was a good knowledge of relevant legislation, but most officers felt they did not have sufficient training in mental illness, and were keen for more.i

Caveat: The demographic details of the sample surveyed are not reported.

i. Carey SJ. Police officers knoweldge of, and attitutes towards mental illness in southwest Scotland. Scottish Medical Journal 2001; 46: 041-042
(Type IV evidence – questionnaire study of 210 police officers in Scotland.)


7.14e Cooperation and improved communication between secure institutions and community teams, with support from district forensic community teams, may help maintain mentally disordered offenders (MDOs) in mainstream services. Results show that all teams experienced difficulties in supporting MDOs. Key workers' proficiency levels did not match demands. Drug- and alcohol-related behaviour problems were significant for all but learning disabilities client groups (93% for drug and alcohol teams; 66% for mental health teams; 66% for probabtion teams; 8% learning disability teams). Psychiatric supervision, day services, and accommodation facilities were common service deficiencies.i

Caveat: The response rate for the surveys was 52%.

i. Vaughan PJ, Pullen N, Kelly M. Services for mentally disordered offenders in community psychiatry teams. Journal of Forensic Psychiatry 2000; 11(3): 571-586
(Type IV evidence – survey study of statutory community teams providing a service to mentally disordered offenders (MDOs) in Wessex. Semi-structured interviews were conducted with community team managers (n=85) and team members (n=665) completed questionnaires on MDO work proficiency perceptions, caseload size, MDO client numbers, demographic characteristics, offence history, and support needs.)


7.14f Results data strongly suggest that collaborations between the criminal justice system, the mental health system and the advocacy community plus essential services reduce the inappropriate use of U.S. jails to house persons with acute symptoms of mental illness. Large differences were found across three sites in the proportion of calls that resulted in a specialised response-28% for the Birmingham, Alabama site, 40% for Knoxville and 95% for Memphis, Tennessee. One reason for the differences was the availability in Memphis of a crisis drop-off centre for persons with mental illness that had a no-refusal policy for police cases. All three programmes had relatively low arrest rates when a specialised response was made - 13% for Birmingham, 5% for Knoxville and 2% for Memphis. Birmingham's programme was most likely to resolve an incident on the scene, whereas Knoxville's programme predominantly referred individuals to mental health specialists.i

Caveat: The results of this study may not be generalisable to the UK where police services and response programmes differ.

i. Steadman HJ, Deane MW, Borum R, Morrissey JP. Comparing outcomes of major models of police responses to mental health emergencies. Psychiatric Services 2000; 51(5): 645-649
cgi/reprint/51/5/645 [accessed 29/07/05
(Type IV evidence – comparative, cross-site descriptive study of three sites (Birmingham, Knoxville and Memphis). Records were examined at each site for approximately 100 police dispatch calls for "emotionally disturbed persons" to examine the extent to which the specially trained professionals responded. Records were also examined for 100 incidents at each site that involved a specialised response.)



7.15 Diversion of care and inreach services
7.15a An effective diversion scheme would have to be able to provide a service to numerous and widely spread courts, most with a low level of activity. In four months 298 prisoners were remanded from North Wales. Only 7 courts remanded more than 16 (equivalent to one remand per week); the busiest court remanded 83, equivalent to five per week. In total, 42 prisoners reported a history of mental disorder, records were available on 28 of these, of whom ten were considered to have a mental disorder requiring admission. Only a small number of individuals in this study required diversion from custody (approximately one per week), but in those cases there was significant unmet need.i

i. Jones C, Jones B, Ward S. Mentally disordered offenders: the need for a diversion service in a rural area. British Journal of Forensic Practice 2002; 4(1):19-23
(Type IV evidence – retrospective observational study to determine how many mentally disorderd offenders from North Wales are remanded in custody, and how many require immediate hospitalisation, and could diversion from custody be effectively achieved in this area. Remand prisoners were identified from prison records. Medical records were examined to identify prisoners reporting a history of mental disorder, psychiatric treatment, self-harm or substance abuse.)


7.15b Although multi-disciplinary mental health in-reach teams can clearly help prisoners cope whilst in prison, until the priorities of such teams are clarified and recruitment into mental health professions improves, such services may succeed in doing nothing more than maintaining prisoners on their medication while making no real progress towards rehabilitation. 77.5% of prisoners felt that being accommodated on the wing had helped them to cope with being in prison. Over half of these prisoners said that this was due to the relaxed atmosphere on the wing rather than any services or assistance provided by the staff. Most respondents saw the wing as simply maintaining inmates on their medication and not offering any therapy or assistance with basic living. Many prison officers working on the wing felt that it was not successful in reducing re-offending, particularly as they saw some prisoners again and again and thought that their treatment was unlikely to continue in the community.i

Caveat: Sampling method and method of data collection and analysis have not been reported. The number of prison officers interviewed from other parts of the prison is unknown.

i. Mills A. Mental health in-reach - the way forward for prison? Probation Journal 2002; 49(2): 107-119
(Type IV evidence – qualitative study of 40 prisoners and eight staff members in England on a wing designed for prisoners with special needs or ‘poor copers’ where a multi-disciplinary mental health in-reach team was present. Officers from other parts of the prison were also interviewed.)


7.15c There is a need to develop a more effective outreach service to assist patients with serious mental illness to engage with services immediately following court appearance. Of the patients diverted to inpatient services from the courts, at least one third had lost contact with psychiatric services at 12-months follow-up. For patients referred from the courts to psychiatric community teams or outpatient clinics, less than one third attended their first appointment and of those, almost one third had become disengaged from services before the follow-up 12 months later.i

Caveat: Follow-up rate for defendants at 12 months was 43.8% (n=39). Age of defendants was not reported, although authors state this was collected.

i. Shaw J, Tomenson B, Creed F, Perry A. Loss of contact with psychiatric services in people diverted from the criminal justice system. Journal of Forensic Psychiatry 2001; 12(1): 203-210
(Type IV evidence – 2-year case-note follow-up study. Case-notes of 235 referrals (male n=198) to a magistrates diversion scheme in Manchester between 1993 and 1995 were reviewed. 89 defendants recommended for psychiatric assessment and treatment by court psychiatrists completed a 12-month follow-up concerning psychiatric service attendance and contact details.)


7.15d This paper examines one of the first prison in-reach services that was launched at HMP Leicester early in 2002, and considers the effect these nurses have had on the care of mentally ill adults at the prison. A case study outlining the in-reach team's approach to one of the prison's greatest challenges, self-harm, is also included.i

i. Armitage C, Fitzgerald C, Cheong P. Prison in-reach mental health nursing. Nursing Standard 2003; 17(26): 40-42
(Type V evidence – expert opinion with case study.)



7.16 Care of offenders with mental health needs
Prevalence of mental disorder in offenders

7.16a Worldwide, several million prisoners probably have serious mental disorders, but how well prison services are addressing these problems is not known. 62 surveys from 12 countries included 22790 prisoners (mean age 29 years, 18530 (81%) men, 2568 (26%) of 9776 were violent offenders). 3.7% of men (95% CI 3.3-4.1) had psychotic illnesses, 10% (9-11) major depression, and 65% (61-68) a personality disorder, including 47% (46-48) with antisocial personality disorder. 4.0% of women (3.2-5.1) had psychotic illnesses, 12% (11-14) major depression, and 42% (38-45) a personality disorder, including 21% (19-23) with antisocial personality disorder. Although there was substantial heterogeneity among studies (especially for antisocial personality disorder), only a small proportion was explained by differences in prevalence rates between detainees and sentenced inmates. Prisoners were several times more likely to have psychosis and major depression, and about ten times more likely to have antisocial personality disorder, than the general population.i

One of the included studies was an Office of National Statistics survey carried out in the UK in 1997. The survey found that a large proportion of all prisoners had several mental disorders. Only 1 in 10 or fewer showed no evidence of any of the five disorders considered in the survey (personality disorder, psychosis, neurosis, alcohol misues and drug dependence) and no more than 2 out of 10 in any sample group had only one disorder.ii

i. Fazel S, Danesh J. Serious mental disorder in 23,000 prisoners: a systematic review of 62 surveys. Lancet 2002; 359(9306): 545-550
(Type IV evidence – systematic review of 62 psychiatric surveys from 12 countries (including the UK) based on interviews of unselected prison populations and included diagnoses of psychotic illnesses or major depression within the previous 6-months, or a history of any personality disorder. Literature search to 2001.)
ii. Singleton N, Melttzer H, Gatward R. Psychaitric morbidity among prisoners in England and Wales. London: Stationery Office, 1998
(Type IV evidence – survey ofpsychiatric morbidity among prisoners in England and Wales. Prevalence estimates for different types of prisoners were achieved through interviews with 3142 male sentenced, male remand, and women prisoners from 131 prisons. A 20% sample was chosen to be followed up of which 76% (n=505) completed clinical interviews.)

7.16b In a county population of 800,400, some 30,329 were offenders. More than a third had used a health or social care service during the three-year period; 8.0% were mentally disordered. Those offenders aged 25-64 and who contacted the police more than once were significantly more likely to be mentally disordered. Type of offence was also a relevant variable. The probation service showed broadly similar results. The research has provided for the first time substantive quantitative evidence of the relationship between crime and mental disorder. The results can be used as the basis for further work to target assessment and risk reduction measures at those most at risk.i

i. Keene J, Janacek J, Howell D. Mental health patients in criminal justice populations: needs, treatment and criminal behaviour. Criminal Behaviour and Mental Health. 2003; 13(3): 168-178
(Type IV evidence – observational study using the Tracking Project method to examine the total mental health agency population of an English county and total criminal justice population for the same area using data from prison and probation services and the community mental health trust.)


7.16c Half of the women in the sample reported at least one act of self-harm in their life and 46% reported making a suicide attempt at some time. Lifetime self-harm was associated with a history of harmful drinking and with being a victim of violence, including physical assault, sexual assault and violence from family and friends. Lifetime suicide attempts were associated with reported violence from family or friends. Current high suicide risk was most common among women on remand. Drug dependence and reported violence from family or friends were both more common amongst white women than black/mixed race women. Self-harm and attempted suicide were generally more common among white women, but black/mixed race women dependent on drugs had the highest proportion of women reporting self-harm. There was tentative support for a three-way association between ethnicity, dependence and self-harm; this raises the possibility that drug dependence may be a predictor of self-harm in the black female prison population.

i. Borrill J, Burnett R, Atkins R et al. Patterns of self-harm and attempted suicide among white and black/mixed race female prisoners. Criminal Behaviour and Mental Health 2003; 13: 229-240
(Type IV evidence – structured interviews with 301 women (190 white, 111 black or mixed race) in ten prisons from different parts of England. Measures included the Alcohol Use Disorders Identification Test (AUDIT), the Severity of Dependence Scale (SDS), and section C (suicidality) of the MINI International Neuropsychiatric Interview.)



7.16d Potential mental health problems that required further specialist assessment were identified in 56% of the young offenders assessed. Alcohol was consumed more than twice per week by 68%, with 47% having recently smoked cannabis, and, 11% recently using heroin, methadone or crack cocaine. Use of secondary health-care services was common although contact with primary-care services was less frequent with almost half having no contact with a GP in the past year.i

Caveat: Study uses a small sample size.

i. Stallard P, Thomason J, Churchyard S. The mental health of young people attending a Youth Offending Team: a descriptive study. Journal of Adolescence 2003; 26(1): 33-43
(Type IV evidence – small-scale survey of 38 young people (aged 10-17 years; 82% males) attending a community Youth Offending Team (YOT) in the UK. Data were collected from self-completed questionnaires, YOT-completed assessment forms and mental health assessments.)


Mental health promotion and prevention of mental illness in prisons

7.16e A preliminary evaluation suggests that Project Link, a university-led consortium of 5 community agencies in New York that spans healthcare, social service, and criminal justice systems, may be effective in reducing recidivism and in improving community adjustment among severely mentally ill patients with histories of arrest and incarceration. Compared to the year before admission to Project Link, mean (SD) yearly jail days per patient dropped from 107.7 (133.5) to 46.4 (83.7) (z=2.6, p<0.01) and mean yearly hospital days dropped from 115.9 (133.5) to 7.4 (17.7) (z=4.3, p<0.001). Significant reductions were also noted in average number of arrests per patient (z=2.9, p<0.005) and in average number of incarcerations and hospitalisations per patient (z<2.7 and p<0.01 ). Mean (SD) Multnomah Community Ability Scale (MCAS) scores improved from 51.5 (7.6) to 61.5 (8.6) (z=5.3, p<.001).i

Caveat: The mirror image design of this study fails to control for age and time effects, and other confounding factors that could have influenced the results. The study sample size was small.

i. Lamberti J, Weisman S, Schwarzkopf RL, et al. The mentally ill in jails and prisons: towards an integrated model of prevention. Psychiatric Quarterly 2001; 72(1): 63-77
(Type III evidence – before and after evaluation study. Service utilisation data were collected for 41 patients (mean age 35.2 years; 78% male), who had completed one year in a programme featuring a mobile treatment team, during the year before admission and the year after admission. Data sources included hospital charts, jail, probation and parole records, and patient functioning assessments.)


7.16f Written materials, while playing an important role in a campaign, are no substitute for clear guidance to members of staff on their role. In health promotion campaigns they must be based on an analysis of need. They cannot be expected to act as ‘stand alone initiatives’ and should be used in well-defined situations, and where possible supported by personal contact. There was a lack of clarity concerning the role and place of the materials in developing or supporting existing policy and health promotion activities. Aims and objectives were not documented and no relationship to any underlying model or theoretical design was made. It was found that a number of individuals in the Health Education Authority (HEA) and the Prison Service had clear ideas about the aim of the project, but not of the materials nor the role they were expected to play. Senior staff in both organisations saw the project as a stepping stone to further collaboration.i ii

i. Caraher M, Bird L, Hayton P. Evaluation of a campaign to promote mental health in young offender institutions: problems and lessons for future practice. Health Education Journal 2000; 59(3): 211-227
ii. Caraher M, Hayton P, Bird L. Mental health promotion in young offender institutions: evaluating a national initiative for World Mental Health Day. Prison Service Journal 2000; 128: 7-12

(Type IV evidence – qualitative study of a health promotion campaign for young offenders in England. Interviews were conducted with HEA and Prison Service staff and semi-structured, telephone interviews were completed with 24 staff members in young offender institutions. 4 staff focus groups were also conducted in 2 institutions and six young offenders completed paired interviews and a focus group for further discussion.)


7.16g Current policy in England and Wales from 1994 focuses upon suicide awareness rather than prevention and is predicated upon a multi-disciplinary and multi-agency approach to caring for the suicidal. Still, self-inflicted deaths continued to increase during the 1990s. One factor cited as evidence that the revised strategy was not working is the number of prisoners who were not identified as at risk at the time of their death. Although the primary focus of suicide awareness strategies is upon staff-prisoner relationships, it is clear that environmental issues play a role in reduction rates. Reducing the means of suicide may reduce the likelihood that a prisoner going through a temporary crisis may complete a suicide. Staff and standardised training should be revised to build confidence in providing support for prisoners at risk. Although a large body of strategies is emerging, there does not appear to be a simple relationship between strategy and a reduction in suicide.i

See also Sections 7.20 7.21 for suicide risk and prevention strategies

i. McHugh M. Suicide prevention in prisons: policy and practice. British Journal of Forensic Practice 2000; 2(1): 12-16
(Type V evidence – expert opinion.)



Treatment of mentally ill offenders
7.16h This meta-analyses found a higher estimation of effect size (r=0.21) than previous meta-analysis psychological programmes with offenders. In specific terms, the treated groups showed a recidivism rate of 39.5% compared to 60.5% of controls. Some typologies of programmes (especially educational, behavioural and cognitive behavioural strategies) were more effective than the average.i

i. Illescas SR, Sanchez-Meca J, Genoves VG. Psychological programmes with offenders and their effectiveness. Psicothema 2002; 14: 164-173
(Type I evidence – systematic review of 26 studies (including 2 randomised controlled trials) with a total of 574 offenders under the control of the criminal justice system (mean age 25.5 years, majority male). Literature search 1980-1998.)

7.16i The results indicate that positive treatment effects were found for the use of group psychotherapy with incarcerated offenders across all outcomes (institutional adjustment 0.43, p<0.0001; anger 0.45, p<0.0001; anxiety 0.94, p<0.0001, depression 0.57, p<0.01; interpersonal relations 0.36, p<0.05; locus of control 0.64, p<0.001; and self-esteem 0.31, p<0.05). Supplemental analyses were also included to identify factors that contribute to the efficacy of group psychotherapy and indicate that the use of homework exercises resulted in significantly improved outcomes (β=0.77, SE=0.35, p<0.05). Furthermore, participants mandated to treatment did not negatively influence the efficacy of group psychotherapy.i

Caveat: The results reported here were based on an analysis that excluded 1 paper that was an outlier (effect size = 2.87; whereas the effect sizes for the remaining papers ranged from 0.02 to 1.12). The design of the studies included in this meta-analysis have not been reported, therefore this has been graded as type III evidence.

i. Morgan RD, Flora DB. Group psychotherapy with incarcerated offenders: a research synthesis. Group Dynamics: Theory, Research, & Practice 2002; 6(3): 203-218
(Type III evidence – meta-analysis of 26 empirical stuides to evaluate the efficacy of group psychotherapy with incarcerated offenders. Literature search date not reported.)


7.16j Successful implementation of the Care Programme Approach (CPA) for all prisoners who meet enhanced CPA criteria is likely to have significant resource implications, both for mental health teams working within prisons and local psychiatric services. Of the 91 prisoners found to fulfill criteria for enhanced CPA, the majority (77%) had a diagnosis of schizophrenia, schizoaffective or delusional disorder, and 58% required transfer to a psychiatric hospital. Of those who required hospital treatment, 75% needed conditions of high- or medium-security.i

Caveat: The criteria for study entry could have introduced bias. Participant characteristics have not been reported.

See also Section 7.1

i. Pyszora N. Implementation of the Care Programme Approach in prison. Psychiatric Bulletin 2003; 27(5): 173-6
(Type IV evidence – 1-year, medical case note study of prisoners detained at a high-security prison in London. 91 cases who had been in contact with the psychiatric team or with the forensic mental health nurse were retrospectively examined to collect data regarding diagnosis, previous contact with psychiatric services, history of substance misuse, nature of index offence and treatment outcomes.)



7.16k The quality of services for mentally ill prisoners fell far below the standards in the NHS. Patients' lives were unacceptably restricted and therapy limited. The present policy dividing inpatient care of mentally disordered prisoners between the prison service and the NHS needs reconsideration. The 13 prisons had 348 beds, 20% of all beds in prisons. Inpatient units had between 3 and 75 beds. No doctor in charge of inpatients had completed specialist psychiatric training. 24% of nursing staff had mental health training; 32% were non-nursing trained healthcare officers. Only one prison had occupational therapy input; 2 had input from a clinical psychologist. Most patients were unlocked for about 3.5 hours a day and none for more than 9 hours a day. 4 prisons provided statistics on the use of seclusion. The average length of an episode of seclusion was 50 hours.i

i. Reed JL, Lyne M. Inpatient care of mentally ill people in prison: results of a year's programme of semistructured inspections. British Medical Journal 2000; 320(7241): 1031-1034
reprint/320/7241/1031 [accessed 29/07/05
(Type IV evidence – semistructured inspections of 13 prisons with inpatient beds in England and Wales subject to the prison inspectorate's routine inspection programme during 1997-8. Inspections lasted 2 to 4 days and were conducted by a medical inspector, a nursing inspector, a professional standards inspector from the Royal Pharmaceutical Society and a dentist from the Dental Practice Board.)


7.16l In this paper, the authors present an overview of the literature regarding the effectiveness of psychological approaches to offender rehabilitation and discuss how the research literature has helped to begin to define best practice in this area. Five principles for rehabilitation: risk, need, responsivity, professional discretion, and programme integrity are highlighted. Finally, the implications of this work for psychologists working with offenders are discussed.i

i. Day A, Howells K. Psychological treatments for rehabilitating offenders: evidence-based practice comes of age. Australian Psychologist 2002; 37(1): 39-47
(Type V evidence – expert opinion.)


Responding to the needs of mentally ill offenders

7.16m Link Workers provide large amounts of practical help in addressing housing problems for people with mental illness and multiple needs who have a history of offending. 44% of clients coming out of prison were given help to make one or more housing applications. In 21% of cases they helped clients to make more than one application. For Housing Benefit, two-thirds of those who successfully applied for this benefit following release did so with the help of a Link Worker. Almost 1 in 4 (24%) of the sample experienced an improvement in their housing tenure following release from prison and while a Revolving Doors Agency client.i

Caveat: The method of quantatative data analysis has not been reported.

i. Revolving Door Agency. ‘Where do they go?’ Mental health, housing and leaving prison. London: Revolving Door Agency, 2002
(Type IV evidence – questionnaire survey and case series study. Data were collected for 101 prisoners in Ealing, serving less than 12-months, before, during and after their prison terms. Questionnaires analysed receipt of benefit, problems with neighbours and accruals of rent arrears.)


7.16n Little difference was found between those in penal and welfare settings except that the "penal group" were much more likely to have high levels of violent behaviour and to have had more changes of placement. The needs for mental health care greatly outstripped supply. As part of this overall neglect, 11 of 15 young people with serious mental illnesses and all 13 who had suffered sexual abuse in the sample were not receiving appropriate treatment. A tentative estimate of the size of the problem in the region yielded a rate of around 11.4 per million with very severe disorder but this is probably an under estimate. The effectiveness of treatment for the problems of these young people is discussed and a possible structure for a service is explored.i

i. Nicol R, Stretch D, Whitney I, et al. Mental health needs and services for severely troubled and troubling young people including young offenders in an N.H.S. Region. Journal of Adolescence 2000; 23(2): 243-261
(Type IV evidence – observational survey study of 116 young people in Trent (penal setting n=57, welfare setting n=59; aged 13-18 years; 86.2% male). A standard psychiatric interview and a needs met/not met assessment was conducted for each participant.)

7.16o Findings show that female mentally ill offenders differ descriptively from their male counterparts. They are younger, more likely to have a history of engagement with social services, and report more trauma. 74% of the women released had a history of receiving Department of Mental Health Services. For men the percentage was 55%. Female mentally ill offenders were most likely to be engaged in community mental health treatment (63%) once released from incarceration for 3-months or more. This is followed in decreasing frequency to being lost to follow-up (16%), immediately hospitalised or ‘stepped-down’ to in-patient hospitals at time of prison release (13%) or recidivating to the hospital (6%) or prison (3%). As a third of the men and women anticipated homelessness, housing is a service priority for many clients. As nearly 10% of the women attempted to adapt to the community only to recidivate, these women would seem to be an especially important target for intervention.i

Caveat: The results of this study may have limited generalisability to a UK setting.

i. Hartwell S. Female mentally ill offenders and their community reintegration needs. An initial examination. International Journal of Law & Psychiatry 2001; 24(1): 1-11
(Type IV evidence – data analysis of 462 mentally ill offenders in Massachusetts (women n=76, men n=386) 3-months prior to release from prison and tracked for up to 3-months post prison release. Data were collated from all client contacts made and recorded by the Forensic Transition Team programme, providing transitional services for mentally ill offenders. Demographic, clinical, service and engagement/recidivism outcome information was analysed.)



National Service Framework: key action 39

Raising the standard. Cardiff: Welsh Assembly Government, October 2005

The needs of vulnerable children and young people whose parents / guardiands have mental health problems are to be considered carefully. There is to be careful planning to ensure their needs are fully taken into account especially in situations where they are acting as carers.[Key action 39 paragraph 28.2]

How can liaison with Child and Adolescent Mental Health Services (CAMHS) be enhanced?
What are the needs of young people whose parents/guardiands have mental health problems, and how can their needs be met?

The Statements The Evidence
7.17 Communication/ liaison with Child and Adolescent Mental Health Services

7.17a To create, enhance and extend the working relationships between GPs and Child and Adolescent Mental Health Services (CAMHS) requires strategies for change in both sectors. There was a two-fold increase in communication between the two sectors and a 120% increase in shared care arrangements. There was a 144% improvement in GPs' perceptions of the helpfulness of public CAMHS (the difference between GPs ratings before and after project implementation was significant, χ2 =5.803, df=2, p=.05). Implications for future collaboration are discussed. The results of this pilot study are encouraging and warrant replication and validation using standardised instruments.i

Caveat: There was a 31% response rate to the GP postal survey before project implementation and a 32% response rate post-implementation. The results of this study may have limited applicability to a UK setting.

i. Mildred H, Brann P, Luk ES, Fisher S. Collaboration between general practitioners and a child and adolescent mental health service. Australian Family Physician. 2000; 29(2): 177-181
(Type III evidence – before and after study in Victoria, Australia . 23 patients were interviewed by case managers before the implementation of a 12-month project to introduce preliminary steps to maintaining collaborative relationships between child and adolescent mental health services and GPs. 32 case managers completed semi-structured interviews and 208 GPs completed postal surveys before and after implementation.)


7.17b Parents felt more confident in managing their children (pre-intervention mean score 5.17; mean post-intervention score 6.33), and identified a positive change in their children’s behaviour, following the Parent Link group intervention (parental Strengths and Difficulties Questionnaire mean pre-intervention score 24.14; mean post-intervention was 19.71). Further modified Life Skills Groups are planned to run again within the Department of Child and Adolescent Psychiatry, Hounslow, and additional parenting groups are being devised.i

Caveat: Very small sample size.

i. Somerville K, Karwatzki E, Simms A. A collaborative venture between a child and adolescent mental health service and the voluntary sector to address a waiting list crisis. Clinical Psychology Forum 2000; 140: 36-40
(Type III evidence – pilot before and after study of 7 families in the UK (children aged between 10-11 years; 2 male, 5 female) attending 12-weekly therapeutic group sessions. Parents attended the Parent-Link group while children attended a separate closed therapeutic life skills group (LSG).)

7.17c From a resource dependency perspective, findings suggest that coordination is facilitated when interorganisational relationships fulfill both the internal agency needs for goal attainment and the external needs for exerting control over the larger policy and programme environment. Greater coordination of activities was significantly associated with dyads that (a) helped each other attain individual agency goals (p<0.001), (b) were influential in shaping mental health policy and programmes (p<0.01), (c) maintained resource linkages over time (p<0.001), and (d) operated in the same service sector (p=0.01).i

Caveat: Data analysed in this study are drawn from 2 waves of data collection in 1991 and 1993.

i. Rivard JC. Morrissey JP. Factors associated with interagency coordination in a child mental health service system demonstration. Administration & Policy in Mental Health 2003; 30(5): 397-415
(Type IV evidence – multiple regression analysis that examined factors associated with coordination between 63 agencies in America during a child mental health service demonstration.)


7.17d Communication problems were identified more frequently between child care workers and adult psychiatrists than between other groups. Communication between general practitioners and child-care workers was also more likely to be described as problematic. While there was some support amongst practitioners for child-care workers to assume a coordinating or lead role in such cases, this support was not overwhelming, and reflected professional interests and alliances. The mothers themselves valued support from professionals whom they felt were 'there for them' and whom they could trust. There was evidence from the responses of child-care social workers that they lacked the capacity to fill this role in relation to parents and their statutory child-care responsibilities may make it particularly difficult for them to do so. The authors recommend that a dyad of workers from the child-care and community mental health services should share the coordinating key worker role in such cases.i

Caveat: There was a 50.5% response rate to the postal survey. The total number of practitioners surveyed is unclear. The sampling method and methods of data collection and analysis have not been reported.

i. Stanley N, Penhale B, Riordan D, Barbour RS, Holden S. Working on the interface: identifying professional responses to families with mental health and child-care needs. Health & Social Care in the Community 2003; 11(3): 208-218
(Type IV evidence – quantatative and qualitative study in North-East England. 500 practitioners with relevant mental health or child protection experience, across a range of different disciplines, completed a semi-structured postal survey.
The views of 11 mothers with severe mental health problems whose children had been subject to a child protection case conference were also interviewed.)


7.17e The four most reported themes were the following: choosing between the Mental Health Act and the Children Act; general issues around consent to treatment; issues with social services departments; and the stigma associated with using the Mental Health Act. The range of themes identified from this survey have served to focus the evaluation of the use of the Children Act and the Mental Health Act in Children and Adolescents in Psychiatric Settings and have informed the design of subsequent data collection tools.i

Caveat: The response rate to participate in the survey was only 51%.

i. Mears A, Worrall A. A survey of psychiatrists’ views of the use of the Children Act and the Mental Health Act in children and adolescents with mental health problems. Psychiatric Bulletin 2001; 25(8): 304-306
(Type IV evidence – UK survey study. 505 members of the Child and Adolescent Faculty of the Royal College of Psychiatrists were asked to complete a single question questionnaire.)



7.18 Needs of children with parents/guardians with mental health problems

7.18a A Royal College of Psychiatrists report is available, providing a practical summary on how psychiatrists can help in a situation where people with a psychiatric disorder or who abuse drugs or alcohol also have child care responsibilities or contact with dependent children.i

i. Royal College of Psychiatrists. Patients as parents: addressing the needs including the safety, of children whose parents have mental illness. London: Royal College of Psychiatrists, June 2002: Council Report CR105
cr/council/cr105.pdf [accessed 29/07/05
(Type V evidence – expert opinion.)

7.18b Location of child mental health services may be less important than the range of services that they provide, which should include effective treatment for parents' mental health problems. Intention to treat analyses showed no significant differences between the community and hospital based groups on any of the outcome measures, or on costs. Parental depression was common and predicted the child's outcome.i

i. Harrington R, Peters S, Green J, Byford S, Woods J, McGowan R. Randomised comparison of the effectiveness and costs of community and hospital based mental health services for children with behavioural disorders. British Medical Journal 2000; 321(7268): 1047-1050
eprint/321/7268/1047 [accessed 29/07/05
(Type II evidence – randomised controlled trial of 141 parents of children (mean age 6.9 years, 79% boys) with behavioural disorder. Subjects were allocated to routine interventions for behavioural disorder in a children’s hospital or in a community setting in the North of England. 12 months follow-up.)

7.18c The needs of children must be considered when depressed mothers are being treated in primary care. The mothers who were prescribed antidepressants (cases) were more depressed than the other mothers (18.7 versus 7.0, p=0.00), and their children had more dysfunctional symptoms (mean total Strength and Difficulties Questionnaire (SDQ) for cases =11.3 SD 6.3, versus controls =8.2 SD 6.4; F=5.7, p=0.02).i

Caveat: Sample sizes in the two groups were small. The interviewer was not blind to the mothers’ status.

i. Hartley K, Phelan M. The needs of children of depressed mothers in primary care. Family Practice 2003; 20(4): 390-392
(Type IV evidence – case-control study set in 3 primary care teams in West London. 30 mothers being prescribed antidepressants by their GP and 30 mothers not being prescribed antidepressants were interviewed, and comparisons made between their children.)

7.18d There is an urgent need for the psychiatric services to initiate parental issues in programmes for treatment and rehabilitation to ensure that the specific needs of minor children are met. Results over the years investigated showed the same proportion of patients admitted to hospital who were also parents to minor children (24% in 1986, 25% in 1991 and 35% in 1997), and a decreasing proportion of patients who had the custody of their children (89% in 1986, 76% in 1991 and 64% in 1997). Female patients were more often a parent (75%, c2=21.14, p=0.0000) and also more often had the custody of the children (86%, c2=22.94, p=0.0000). The majority of the children had needs for support caused by their parent's illness (55%) and these needs were met in half of the cases (54%).i

i. Ostman M, Hansson L. Children in families with a severely mentally ill member. Prevalence and needs for support. Social Psychiatry & Psychiatric Epidemiology 2002; 37(5): 243-248
(Type IV evidence – observational longitudinal study of relatives of both voluntarily and compulsorily admitted patients to acute psychiatric wards, who had participated in a multi-centre study of the quality of the mental health services in Sweden performed in 1986, 1991 and 1997. 422 participants (mean age 40 years, 54% women) completed semi-structured questionnaires.)


7.18e Depression among children of depressed mothers is especially likely to occur in the context of - and perhaps, result from – difficulties in their interpersonal skills and perceptions of others. As predicted, after controlling for current symptoms and family social status variables, depressed offspring of depressed mothers displayed significantly more negative interpersonal behaviours (F1, 762=16.52, p<0.0001) and cognitions (F1, 740=3.00, p<0.08) compared with depressed offspring of nondepressed mothers, but they did not differ on academic performance. Within the depressed groups, children of depressed mothers had significantly elevated rates of interpersonal and conflict events compared to offspring of nondepressed mothers, t(102)=2.01, p<0.05 and t(102)=1.86, p<0.05, respectively. One of the most problematic interpersonal domains for the depressed youth of depressed mothers was quality of family relationships, with the groups differing significantly, t(102)=2.73, p<0.004.i

i. Hammen C,Brennan PA. Depressed adolescents of depressed and nondepressed mothers: tests of an interpersonal impairment hypothesis. Journal of Consulting & Clinical Psychology 2001; 69(2): 284-294
(Type IV evidence – 15-year follow-up of a birth cohort study in Australia of prenatal factors affecting child development. 65 (69.2% female) depressed offspring of women (mean age 41 years) with histories of a major depressive episode or dysthymia were compared with 45 depressed (80% female) offspring of never-depressed women (mean age 41 years)



7.18f The problem of mismatch between need and utilisation must be addressed. Mothers had been admitted to psychiatric hospital 1-9 times (mean=3.8 admisions) since their child's birth, with length of hospitalisation between 1 week and 9 months. 11 children had a lifetime psychiatric diagnosis from at least one source, and 10 had a current diagnosis. In terms of comorbidity, 11 had more than one lifetime diagnosis, and 9 had more than one current diagnosis. The common diagnoses were mood, anxiety, and behavioural disorders. Service utilisation was low, with only 2 children (siblings) in current treatment with a child mental health service. Interviews suggest that a large number of vulnerable children with significant psychiatric disorders were not receiving help from local services and were not being referred at all.i

Caveat: Mothers were offered £10 vouchers for study participation. The follow-up rate was 41.3%.

i. Singer J, Tang S, Berelowitz M. Needs assessment in the children of parents with major psychiatric illnesses. In: Reder P, McClure M, et al, eds. Family matters: Interfaces between child and adult mental health. New York: Routledge 2000; 192-209
(Type IV evidence – audit study of 29 mothers (aged 17-55 years) with major mental illness from the Royal Free Hospital catchment area in London. Mothers and children (n=55 children, mean age 10.9 years) completed diagnostic interviews, self-report behavioural questionnaires and mothers answered questions about service utilisation. Children also completed semi-structured interviews.)


7.18g This overview provides an analysis and assessment of the literature that deals with children of a parent who has a mental illness. It argues that children's perspectives about living with a parent who has a mental illness have not been taken into consideration. A survey of the literature indicates that it can be divided into sections that include: the family context of the child, risks associated with the child's stage of growth and development, characteristics associated with resilience, and existing interventions. The authors propose a programme of research that addresses the issues raised in the analysis.i

i. Mordoch E,Hall WA. Children living with a parent who has a mental illness: a critical analysis of the literature and research implications. Archives of Psychiatric Nursing 2002; 16(5): 208-216
(Type V evidence – expert opinion.)


7.18h Services urgently need to involve adolescents in the development of mental health services, but before this can be done, there is a need to establish the views of adolescents. On the whole, children and adolescents appear to be infrequently asked about their views of health services and whether these correspond to their perceived needs and requirements.i

i. Dogra N. Adolescent perspectives on the provision of services for their mental health needs. European Child & Adolescent Psychiatry 2000; 9: 70-73.
(Type V evidence – expert opinion.)


7.18i The document ‘Hidden Harm’ focussing on children in the UK with a parent, parents or other guardian whose drug use has serious negative consequences for themselves and those around them, is available electronically. The report outlines several recommendations on areas such as;

  • The impact of parental problem drug use on children
  • The voices of children and their parents
  • Surveys of specialist drug agencies, maternity units
  • and social work services
  • The legal framework and child protection
  • arrangements
  • The practicalities of protecting and supporting the

children of problem drug users.i

i. The Advisory Council on the Misuse of Drugs. Hidden Harm: Responding to the needs of children of problem drug users. June 2003  [accessed 1/11/05]
(Type V evidence – inquiry carried out by the Advisory Council on the Misuse of Drugs. The working group consisted of members from diverse backgrounds and disciplines, predominantly in the fields of drug use and children’s services. It carried out extensive reviews of published research and reports, commissioned analyses of existing data and national surveys and took evidence from a wide range of expert witnesses.)


Young carers

7.18j There was a good deal of variation in the situation of young carers and in their responses to it. The emotional impact of the young carer's role was what came out most strongly. Most young people were not seeking to be 'rescued' from their role and what they saw as their responsibilities towards their family. What they were seeking was (a) good support services, (b) information and dialogue with service providers, and (c) social and emotional support for themselves. There was no evidence that the broad pattern of need is any different in Wales from elsewhere, although it is likely that the rural nature of much of Wales and the extent of poverty will exacerbate the situation of many young carers and create additional challenges for service providers. Where services were provided, young carers often felt excluded by not being informed, consulted, or treated with respect.i

Agencies must make more concerted efforts to identify young carers in their locality, based on a definition that can be agreed on a Wales wide basis. Intra/inter agency working arrangements must be strengthened if a strategic approach to the identification and support of young carers and their families is to develop. Practitioners must devise appropriate assessment tools to use with young carers and their families and, following an initial assessment, needs must be reviewed on a regular basis. There was no clear framework as to who was responsible for young carers within local authorities. Practitioners lacked an explicit framework for assessing young carers' needs. The extent of multidisciplinary training and working in respect of young carers varied across local authorities. Local Education Authorities, in particular, were largely unaware of the existence of young carers in their area and their needs, circumstances and support requirements. Young Carers Projects offered an invaluable source of support to young carers and their families and they were also an important provider of information. Despite the high level of commitment to supporting young carers there was a significant gap between policy and practice.ii

i. Thomas N, Stainton T, Cheung W et al. A study of young carers in Wales: perspectives of children and young people. Report for Wales Office of Research and Development for Health and Social Care. Swansea: Centre for Applied Social Studies, University of Wales, 2001
(Type IV evidence – qualitative study of 21 primary young carers (mean age 14 years, 62% girls) who completed semi-structured interviews and 6 young carers who participated in initial focus groups.)
ii. Seddon D, Jones K, Hill J, Robinson C. A study of young carers in Wales. Report for Wales Office of Research and Development for Health and Social Care. Bangor: University of
Wales, 2001
(Type IV evidence – qualitative study. 67 key personnel from Local Education Authorities, Social Services, Local Health Groups and voluntary agencies, involved in supporting young carers, completed structured telephone interviews and 76 young carers (mean age 13 years) attended focus groups.)



7.18n Early caregiving is not associated with poor mental health in adulthood for many young caregivers. However, some individuals do appear at risk of depression in adulthood. Results showed that the sample reported more positive mental health than negative mental health, (t(23)=3.86, p<0.001) though 42% had high depressive scores on the total CES-D. More individuals perceived over-protection from a father was associated with higher scores on the CES-D (total score) (r(20)=0.52, p=0.02). Individuals who reported fathers as too protective reported less current positive mental health.i

Caveat: The study had a small sample size.

i. Shifren K. Kachorek LV. Does early caregiving matter? The effects on young caregivers' adult mental health. International Journal of Behavioural Development 2003; 27(4): 338-346
(Type IV evidence – cross-sectional survey of 28 individuals in the US (21 to 58 years) who completed brief phone interviews and then completed mailed questionnaires on their early caregiving experiences, mental health, and early parent-child relations. All participants had provided primary caregiving assistance (i.e., bathing, feeding, etc.) for a parent or adult relative when the caregiver was under 21 years of age.)


National Service Framework: key action 41

Raising the standard. Cardiff: Welsh Assembly Government, October 2005

Suicide prevention is a priority for services. It is to be addressed by delivering high quality and responsive effective evidence based care using relevant NICE guidelines and the recommendations of National Confidential Inquiry into Homicide and Suicides Safety First. [Key action 41 paragraph 32.4]

What are the risks for suicide and what are effective  interventions for preventing suicide?

What are the recommendations from guidelines and The National Confidential Inquiry into Homicide and Suicides Safety First for preventing suicide? 

See also Sections 7.147.16 for care of the mentally ill and suicide prevention in prisons

The Statements The Evidence
7.19 Suicide prevention

7.19a There was insufficient evidence to make any firm recommendations about the most effective form of clinical intervention, psychosocial treatment, or pharmacological treatment for patients who deliberately self‑harmed themselves. Evidence from reviews, with a good methodological quality rating, suggests that some types of psychosocial and pharmacological treatments including problem­-solving therapy, provision of a card for emergency contact, flupenthixol treatment and dialectical behavioural therapy appear promising in reducing rates of repeated self‑harm among suicide attempters. There was uncertainty and insufficient evidence as to the safety and effectiveness of school‑based preventive programmes for adolescents. The programmes directed to the at‑risk students appeared promising in terms of reduction in suicidal risk behaviours and enhancement of protective factors. Suicide is complex and multifaceted and therefore requires a combination of prevention/ treatment strategies to achieve reduction in suicide rates.i

Caveat: Unpublished research was not sought.

i. Guo B, Scott A, Bowker S. Suicide prevention strategies: evidence from systematic reviews. Health Technology Assessment: 28. Edmonton: Alberta Heritage Foundation for Medical Research, 2003.
(Type I evidence – systematic review of 10 systematic reviews to evaluate the effectiveness of suicide prevention interventions covering a wide range of prevention strategies. Literature search to 2002.)



7.19b Overall, the evidence suggests that no firm conclusions can be reached on the efficacy of a variety of different kinds of follow-up, largely due to the small size of most trials and the variety of interventions and non-standardisation of ‘standard’ care, making it difficult to perform meta-analyses. There is some evidence to suggest that cognitive-behavioural strategies may reduce repeat suicide attempts but it is unknown which sub-groups of patients would most benefit; also any positive effect seen diminishes with follow-up periods of longer than 6-12 months.i

Caveat: Unpublished research was not sought.

i. Hall K, Day P. Suicide prevention topic 1: What kind of follow-up is needed to reduce the risk of repeated suicide attempts/suicide? NZHTA Report 2002. Christchurch: New Zealand Health Technology Assessment (NZHTA), 2002 [accessed 29/07/05]
(Type I evidence – systematic review of 3 systematic reviews and meta-analyses and 10 randomised controlled trials examining suicidal behaviour or suicide as a formal outcome measure in relationship to interventions (follow-up) put in place after initial treatment for a suicide attempt. Literature search 1990-2001.)


7.19c Overall, the evidence suggests that no firm conclusions can be reached on the efficacy of different crisis interventions, largely due to the limited number of trials that have taken place and to the small size of most of these trials. The variety of interventions and non‑standardisation of `standard' care makes any kind of comparison of interventions difficult. There is very little evidence to suggest that crisis intervention as opposed to 'standard care' reduces repeat suicide attempts. From the evidence that was found, it would appear that specialist telephone help lines may be of some help in decreasing suicidal urgency in the short‑term, but evidence is lacking as regards to the long‑term benefits.i

Caveat: This review has been limited to the published academic literature.

See also Section 6.5 and 7.5 for further information on crisis interventions

i. Day P, Dawson S. Suicide prevention topic 2: What is the efficacy of crisis interventions? NZHTA Report 2002. Christchurch: New Zealand Health Technology Assessment (NZHTA), 2002 [accessed 29/07/05]
(Type I evidence – systematic review of 8 studies (1 systematic review, 3 RCTs, and 4 before and after studies) to examine the efficacy of crisis interventions for persons presenting following a suicide attempt, expressing suicidal ideation and suicide threat. Primary outcomes considered were the impact during follow-up on repeat presentations for suicidality, repeat suicide attempts and mortality from suicide. Literature search 1990-2002.)


7.19d Overall, the evidence suggests that no firm conclusions can be reached on the efficacy of different outcomes associated with different triage models used in people presenting following suicide attempt, largely due to the limited number of trials that have taken place and to the small size of most of these trials. The variety of interventions and non‑standardisation of standard care makes comparison of outcomes associated with different triage models difficult. There is little evidence to suggest that different triage methods produce different outcomes in people presenting following suicidal crisis. From the evidence that was found, it would appear that psychosocial assessment to identify high‑risk patients may be of some help in reducing repeat suicide attempts. However, there is no standardised psychosocial assessment for suicidal patients which makes assessment of triage methods and outcomes complicated.i

Caveat: This review has been limited to the published academic literature.

i. Dawson S. Suicide prevention topic 4: Are different triage models associated with different outcomes in people presenting following suicidal ideation/threat/attempt? Christchurch: New Zealand Health Technology Assessment (NZHTA), 2002. [accessed 29/07/05]
(Type I evidence – systematic review of 8 studies (1 RCT, 1 descriptive and 6 case-control studies) to examine the efficacy of different outcomes associated with different triage models used for persons presenting following a suicide attempt, expressing suicidal ideation and suicide threat. Primary outcomes measured the impact during follow-up on repeat presentations for suicidality, repeat suicide attempts and mortality from suicide. Literature search to 1990-2002.)


7.19e There was no research identified addressing the review question other than through expert opinion and the reporting of small case studies and these only indirectly. Literature identified in the search strategy of peripheral relevance pertained to evaluating follow-up and treatment interventions, compliance with follow-up, liability for premature discharge or whether health contact predict future suicide attempts. However, there was no research identified investigating whether having a discharge plan itself affects suicidality outcomes. As efficacy was not demonstrated, it was therefore not possible to discuss what should be included in discharge plans.i

Caveat: This review has been limited to the published academic literature and to emergency department or tertiary mental health settings.

i. Broadstock M. Suicide prevention topic 12: What is the efficacy of discharge planning protocols, i.e., managing the transition from hospital to community? What should be included in the plan? NZHTA Report 2002. Christchurch: New Zealand Health Technology Assessment (NZHTA), 2002. [accessed 29/07/05]
(Type I evidence – systematic review to evaluate the efficacy of discharge planning protocols in managing the transition from the hospital (emergency department or tertiary mental health setting) to the community of patients presenting for suicidality. Literature search 1990-2002.)



7.19f There was no relevant literature identified which directly addressed crisis containment drug treatments for reducing suicidality and met the selection criteria. i

Caveat: Unpublished research was not sought.

i. Broadstock M. Suicide prevention topic 14: Are there any crisis containment drug treatments that have been shown to be useful for reducing suicidality in short-term crises? NZHTA Report 2002. Christchurch: New Zealand Health Technology Assessment (NZHTA), 2002 [accessed 29/07/05]
(Type I evidence – systematic review of the use of rapid-onset drugs for short-term crisis-containment of patients presenting with suicidality in emergency departments or tertiary mental health settings. Literature search 1990-2002.)

7.19g Hallucinations and suicide attempt before inclusion in the study were the most significant predictors of suicide attempt in the follow-up period. During the 1-year follow-up period, 11% attempted suicide. This was associated with female gender, hopelessness, hallucinations (OR 2.00, 95% CI 1.34-2.97) and suicide attempt reported at baseline (χ2 10.21, p<0.001), with the 2 latter variables being the only significant ones in the final multivariate model. The integrated treatment reduced hopelessness.i

Caveat: Follow-up rate was 66.6%. The results of this study are based on a 1-year interim analysis. It is unclear whether or not an intention-to-treat analysis was used.

i. Nordentoft M, Jeppesen P, Abel M, Kassow P, Petersen L, Thorup A et al. OPUS study: suicidal behaviour, suicidal ideation and hopelessness among patients with first-episode psychosis. One-year follow-up of a randomised controlled trial. British Journal of Psychiatry 2002; 181(Suppl 43): S98-S106
(Type II evidence – 2-site randomised controlled trial of 341 patients (mean age 27 years; 60.1% males) with a first-episode schizophrenia-spectrum disorder allocated to receive either integrated treatment or treatment as usual. 2-year study period.)


7.19h Clozapine therapy demonstrated superiority to olanzapine therapy in preventing suicide attempts in patients with schizophrenia and schizoaffective disorder at high risk for suicide. Use of clozapine in this population should lead to a significant reduction in suicidal behaviour. Suicidal behaviour was significantly less in patients treated with clozapine versus olanzapine (hazard ratio 0.76, 95% CI 0.58-0.97, p =0.03). Fewer clozapine-treated patients attempted suicide (34 versus 55, p =0.03), required hospitalisations (82 versus 107, p =0.05) or rescue interventions (118 versus 155, p =0.01) to prevent suicide, or required concomitant treatment with antidepressants (221 versus 258, p =0.01) or anxiolytics or soporifics (301 versus 331, p =0.03). Overall, few of these high-risk patients died of suicide during the study (5 clozapine versus 3 olanzapine-treated patients, p =0.73).i

i. Meltzer HY, Alphs L, Green AI, et al. Clozapine treatment for suicidality in schizophrenia: International Suicide Prevention Trial (InterSePT). Archives of General Psychiatry 2003; 60(1): 82-91
(Type II evidence – multicenter, randomised controlled trial of 980 patients (mean age 37.1 years; 61.4% male) with shcizophrenia or shizoaffective disorder who were considered at high risk for suicide assigned to treatment with either clozapine or olanzapine. Primary end points included suicide attempts (including those that led to death), hospitalisations to prevent suicide, and a rating of "much worsening of suicidality" from baseline. 2-year study.)


7.19i A systematic programme of contact with persons who are at risk of suicide and who refuse to remain in the health care system appears to exert a significant preventive influence for at least 2 years. Diminution of the frequency of contact and discontinuation of contact appear to reduce and eventually eliminate this preventive influence. Patients in the contact group had a lower suicide rate in all five years of the study. Formal survival analyses revealed a significantly lower rate in the contact group (p=0.04) for the first 2 years; differences in the rates gradually diminished, and by year 14 no differences between groups were observed.i

i. Motto JA. A randomized controlled trial of postcrisis suicide prevention. Archives of General Psychiatry 2003; 60(1): 82-91
(Type II evidence – unblinded randomised controlled trial of 843 patients (mean age 34 years, 56% women) who had refused ongoing care, assigned to either an intervention group who were contacted by letter at least 4 times a year for 5 years or a control group-who received no further contact. Suicide rates in the contact and no-contact groups were compared. The study was conducted in San Francisco and had a 15-year follow-up.)


7.20 Suicide risk
7.20a Suicide risk in patients with anxiety disorders is higher than previously thought. Patients with anxiety disorders warrant explicit evaluation for suicide risk. Overall, among 20076 participating anxious patients, 12 committed suicide and 28 attempted suicide. The annual suicide risk rate was 193/100000 patients and annual suicide attempt risk was 1350/100000 patients.i

i. Khan A, Leventhal RM, Khan S, Brown WA. Suicide risk in patients with anxiety disorders: a meta-analysis of the FDA database. Journal of Affective Disorders 2002; 68(2-3): 183-190
(Type I evidence – meta-analysis of FDA clinical trial data for pharmacological agents approved for the treatment of anxiety disorders. Data were analysed for incidence of suicides and suicide attempts.)

7.20b After 1-year non-fatal repetition rates were around 15%. The strong connection between self-harm and later suicide lies somewhere between 0.5% and 2% after 1-year and above 5% after 9 years. Suicide risk among self-harm patients is hundreds of times higher than in the general population. Median proportions for repetition 1 year later were: 16% non-fatal and 2% fatal; after more than 9 years, around 7% of patients had died by suicide. UK studies found particularly low rates of subsequent suicide.i

Caveat: Unpublished studies have not been sought.

i. Owens D, Horrocks J, House A. Fatal and non-fatal repetition of self-harm: systematic review. British Journal of Psychiatry 2002; 181: 193-199
(Type I evidence – systematic review of 90 cohort studies and clinical trials where patients had been recruited to a study after attending a general hospital as a result of an episode of non-fatal self harm and reported the propotion that repeated self-harm – fatal or not – for any follow up period of at least one year. Literature search 1970-2001.)



7.20c Overall, the evidence suggests that no firm conclusions can be reached on the characteristics of repeating versus non‑repeating suicidal presenters, which is largely due to the limited number of trials that have taken place and to the small size of most of these trials. There is some evidence to suggest that repeat suicide attempts are more common amongst people with a history of psychiatric contact/illness, psychiatric admission, a history of deliberate self harm/deliberate self poisoning, people who abuse drugs/alcohol, the unemployed, the unmarried and people who expressed threat/left note/plan. More research is needed in this area.i

Caveat: Unpublished research was not sought.

i. Dawson S. Suicide prevention topic 6: What are the characteristics of repeating vs. non-repeating suicidal presenters to Emergency services? NZHTA Report 2002. Christchurch: New Zealand Health Technology Assessment (NZHTA), 2002 [accessed 29/07/05]
(Type IV evidence – systematic review of 12 non-analytical studies (case series and descriptive) examining the characteristics of repeating versus non-repeating suicidal presenters to emergency services. In addition, 5 papers that examined the characteristics of repeating versus non-repeating suicidal presenters out with the emergency services were also assessed. Literature search 1990-2002.)


7.20d No formalised study was identified to either support or refute the assertion that enquiring about suicidal intent does not affect the subsequent rate of attempts. Despite this lack of literature, it is widely and strongly asserted in many professional guidelines concerning the management of suicidal patients that no such risk exists. However, this review failed to identify and establish the evidence for this assertion. Either the evidence exists but has not been identified, or the evidence does not exist in the published literature. Either way the evidence base for this assertion remains unknown.i

Caveat: Unpublished research was not sought.

i. Hall K. Suicide prevention topic 7: Does asking about suicidal ideation increase the likelihood of suicide attempts? NZHTA Report 2002. Christchurch: New Zealand Health Technology Assessment (NZHTA), 2002 [accessed 29/07/05]
(Type I evidence – systematic review to examine whether asking a patient if he/she were considering suicide subsequently increases the risk of that patient either attempting or completing the act. Literature search 1990-2002.)


7.20e The evidence from the appraised literature indicates that clinician competency in making suicide risk assessments, in terms of asking the right questions and recording relevant information, improves when some form of structured psychosocial evaluation is used. It also may indicate that assessments by mental health specialists gain a greater ascertainment of critical information than non‑specialists. The evidence reviewed shows that medical records can provide, or fail to provide, important case information pertaining to the assessment and care of suicidal patients. Medical records provide an important means of communication between care providers. Ideally, these should document a review of previous treatment received, family member concern, relevant suicide risk assessments with each outpatient visit, and risk/benefit assessment of each significant clinical decision.i

Caveat: This review has been limited to the published academic literature.

i. Day P. Suicide prevention topic 8: Is there any evidence regarding the competency of different clinicians to do adequate suicide risk assessments? Christchurch: New Zealand Health Technology Assessment (NZHTA), 2002 [accessed 29/07/05]
(Type I evidence – systematic review of 1 randomised controlled trial and 6 descriptive studies to examine the evidence regarding the competency of different clinicians (psychologists, psychiatrists, nurses, emergency department physicians) to complete adequate suicide risk assessment. Literature search 1990-2002.)



7.20f Seclusion or containment is a procedure used in a variety of settings to manage acute and escalating risk in suicidal patients. It may also involve the use of other forms of restraint and appropriate monitoring. Although the literature search did identify a reasonable number of articles relating to this subject, the vast majority of studies retrieved were based upon expert opinion or dealt with nursing staff or consumer perceptions of the interventions. Studies rarely addressed the subtopics specifically. Although some studies did report the average length of time an individual spent in seclusion, they did not make any assessment of the optimal duration. There was very little literature on seclusion room design. Only 1 study was formally appraised for the review topic. This study was of poor quality and did not use standard case‑control methodology. Whilst of borderline eligibility, it was included given the paucity of data for this topic.i

Caveat: This review has been limited to the published academic literature.

i. Doughty C. Suicide prevention topic 9: What evidence is there about the use of seclusion or containment for patients presenting with suicidal behaviours at emergency departments, tertiary mental health services or inpatient units? Christchurch: New Zealand Health Technology Assessment (NZHTA), 2002[accessed 29/07/05]
(Type III evidence - systematic review of 1 case control study to evaluate the effectiveness of the use of seclusion/containment for crisis management. Literature search to 2002.)


7.21 Recommendations from guidelines and The National Confidential Inquiry into Homicide and Suicides Safety First

7.21a Evidence based guidelines are available for the short-term physical and psychological management and secondary prevention of self-harm.

Recommendations include that all people who have self-harmed should be assessed for risk; this assessment should include identification of the main clinical and demographic features known to be associated with risk of further self-harm and/or suicide, and identification of the key psychological characteristics associated with risk, in particular depression, hopelessness and continuing suicidal intent. Further recommendations regard management in primary care and emergency departments, medical and surgical management, support and advice for people who repeatedly self-harm and psychological, psychosocial and pharmalogical interventions.i

i. National Institute for Clinical Excellence. Self-harm. The short-term physical and psychological management and secondary prevention of self-harm in primary and secondary care. Clinical Guideline 16 London: NICE. July 2004. Review date: July 2008
CG016NICEguideline.pdf [accessed 29/07/05]

(Evidence based guideline with systematic literature search and expert consensus.)


7.21b A broadly based suicide prevention strategy is needed in each country. This should set out what actions should be taken by mental health services as well as other health and social care services. Clinical services should place priortiy for suicide prevention and monitoring on in-patients who are: under non-routine observations; assessed to be at high risk or who are detained and in the first 7 days of admission; at high risk and who are sufficently recovered to allow home leave but whose home circumstances lack support. Priority should also be given for recently discharged patients who are at high risk or who were recently detained, and patients who become non-compliant or who miss service contact while under enhanced CPA (or its equivalent in Scotland Wales and Northern Ireland.

Please refer to the whole document for full statistical results and recommendations regarding inpatients and post-discharge follow-up, care programme approach, staff training, substance misuse, ethnic minorities, criminal justice system and stigma.i

i. The National Confidential Inquiry. Safety first: Five-year report of the National Confidential Inquiry into Suicide and Homicide by People with Mental Illness. London: Department of Health, 2001
[accessed 29/07/05]
(Type IV evidence - findings of the National Confidential Inquiry into Suicide and Homicide by People with Mental Illness based on information received by the Inquiry between 1996 and 2001 and including data for England, Wales, Scotland and Northern Ireland. Data were first
collected on a comprehensive national sample of suicides irrespective of mental health history. Individuals were then identified within the sample who had been in contact with mental health services and clinical data about these individuals collected.)



Contents Home

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