Observation of People with Mental Health Problems

A revision of the 1995 CRAG document Nursing Observation of Acutely Ill Psychiatric Patients in Hospital to bring it in line with current clinical practice and policy terminology.




The Group was established in July 2000 with the following remit:

To develop a framework of practice for clinical teams in acute inpatient units to set and implement patient observation levels in ways that

  • reduce the risk of untoward incidents to both patients and staff

  • combines the practice of observation levels with the proper assessment, treatment and therapeutic engagement with patients

  • maintains a balance between the patient's dignity and the need to ensure that he/she does not come to avoidable harm.


  • to encourage the integration of observation practice into good quality inpatient ward therapeutic care

  • to provide standards for clinical practice, enabling the processes of:

  • audit

  • skill definition

  • training

  • supervision

  • quality improvement to occur.


  • to review the development of policy and practice as it has occurred over the last five years since the publication of Nursing Observation of Acutely Ill Psychiatric Patients in Hospital (1995)

  • to incorporate recent policy developments and practice into a revised good practice guideline to facilitate:

  • clinical governance

  • clinical risk management

  • continuous organisational learning through examination of practice such as critical incident reviews.



Mr David Bertin (Chairman)

Clinical Nurse Manager - Lomond and Argyll Primary Care Trust

Mrs Susan Bishop

Chief Pharmacist - Forth Valley Primary Care Trust

Ms Joan Blackwood

Practice Development Facilitator/Research Associate - Mackinnon Unit, Stobhill Hospital

Dr Keith Brown

Consultant Psychiatrist - Forth Valley Primary Care Trust

Ms Moira Cossar

Clinical Nurse Manager - Crichton Royal Hospital

Dr Tim Dalkin

Consultant Psychiatrist - Royal Edinburgh Hospital

Mr Bob Gillies

Intensive Psychiatric Care Unit - Gartnavel Royal Hospital

Mr Michael Hughes

Ward Manager - State Hospital

Ms Elaine Hunter

Trust Advisor in Occupational Therapy - Royal Edinburgh Hospital

Ms Ruth Lockwood

CRAG Secretariat - Scottish Executive Health Department

Dr John Loudon

Principal Medical Officer - Scottish Executive Health Department

Mr Jamie Malcolm

Nursing Officer - Mental Welfare Commission for Scotland

Ms Jackie Meikle

Ward Manager - Royal Edinburgh Hospital

Mrs Alison Meiklejohn

Head Occupational Therapist - Royal Edinburgh Hospital

Ms Corinna Penrose

Senior Advocacy Worker - Advocacy Matters

Dr Linda Pollock

Nursing Director - Lothian Primary Care NHS Trust

Mr Colin Poolman

Royal College of Nursing Professional Officer - RCN Scottish Board

Mr Robert Samuel

Nursing Adviser - Scottish Executive Health Department

Mr Mark Simpson

Clinical Nurse Manager - Royal Dundee Liff Hospital

Mrs Frances Smith

Director of Nursing and Quality - Clinical Standards Board for Scotland

Ms Lesley Wilkes

Adviser, Mental Health, Scottish Health Advisory Service



To ensure that the work of the group was based upon the experience of the Service three consultation exercises were undertaken:

  • a review of Trust observation protocols

  • a survey on the impact and barriers to implementing the original good practice statement on observation

  • a brief survey of service users' experiences of observation.


In October 2000 Chief Executives of Primary Care Trusts were invited to submit their current protocols to enable the group to develop an understanding of the impact on observation of the 1995 "CRAG Nursing Observation of Acutely Ill Psychiatric Patients in Hospital" report. Comments were invited on the original publication and implementation issues.

The vast majority of protocols refer to three levels of observation and were broadly in line with the suggestions in the original document although different terminology had developed; some policies referred to "close" observation as opposed to "constant" and some referred to "intensive" observation as opposed to "special". Three documents made specific recommendation to a fourth level of observation that sat between "general" and "constant" observation. One reply indicated that this level of observation had been left in due to a resistance from staff to alter this practice.

Recurring themes requiring greater clarity were: guidance on the observation role of other professionals and relatives, training, recording, post-incidence reviews and supervision, risk assessment, patient rights, and the therapeutic nature of observation.


In December 2000, a questionnaire was sent out to Chief Executives of every Primary Care Trust seeking responses from senior nurses in acute wards and IPCUs. A total of 80 responses was received representing every Primary Care Trust. The purpose of the questionnaire was two-fold:

(a) to understand the impact and obstacles to implementing the 1995 nursing observation good practice statement


(b) to ensure that the views and experience of the Service were incorporated into the revised document.

In regard to the three levels of observation recommended, does your service policy broadly fall in line with these?

Virtually all respondents stated that their observation levels corresponded with the CRAG recommendations. 5/15 respondents commented that they used a fourth level of observation that tended to be "general with conditions".

What type of training, if any, has been carried out for staff regularly involved in observation of patients?

Training varies from Trust to Trust; 25% had received some training in observation as part of an induction package; 10% noted some awareness training and two noted
on-going training courses and regular team discussions. Only 6% were required to reach a competence level before being considered competent to carry out observation and 15% reported no training at all.

What particular problems have there been with implementing your observation policy and procedure?

Resource implications. Almost half of the respondents reported staffing problems, in particular low and decreasing numbers of trained staff and untrained bank staff unfamiliar with the ward. This leads to no cover during nurse breaks. Some concern was expressed that staffing restraints led to custodial rather than therapeutic care that compromises patients' privacy and dignity. Low staffing can have a negative impact upon other ward activities if there are high numbers of patients under observation. Others identified environmental issues as particular concerns. 16% reported problems with recalling medical staff to carry out medical reviews/reduce levels, particularly at weekends. Two respondents expressed concern that the observation policy is misused. Concerns were also expressed at the use of other professionals (mainly Occupational Therapists) to observe acutely-ill patients. 11% reported no problems.

What are the current issues and concerns within your own area at the moment regarding observation of patients?

Major problems are resource implications and lack of qualified staff. 14% reported difficulties contacting medical staff to approve lowering of observation levels. Several respondents highlighted issues of risk and risk assessment _ lack of training and lack of confidence, anxieties that individual staff will be blamed/involved in legal actions if problem occurs. Staff fatigue working full time on observation. Seven respondents reported local environmental issues. Only four reported no problems or concerns.

If observation was to be moved from a purely nursing responsibility to one that could be shared among other professionals and potentially relatives, what would your views be on this and what concerns or issues would you like the group to address?

One-fifth reported that shared observation is already happening although not all wards extend this to relatives and carers. A third would welcome future involvement of other professionals, more than half of this group would also welcome some involvement with relatives but cautioned that training and guidelines would have to be provided to ensure observation retained its therapeutic role. Many were concerned about relinquishing responsibility and decision making on behalf of the patient. Only 10% would not welcome the involvement of other professions and relatives stressing that observation was a skilled nursing intervention based upon professional experience.

Any other concerns, questions or information that you feel would influence the group?

There was a great interest in more tailored training to support nursing staff _ topics identified were risk management, symptom identification, team working, record keeping and liability. There was a call for additional research/audit and a request for a sample audit tool to support new guidance. It was noted that returning responsibility to patient can be disruptive _ this should be dealt with sensitively.


Question One: During either your current stay in hospital or on a previous occasion, have you been placed on a "Nursing Observation Level"?

Question Two: Were you given a clear and understandable verbal explanation of:

  • Why you were being observed?

  • How it would be carried out?

  • By whom?

  • For how long?

  • Your rights?

  • Told of any restrictions on your movements?

  • Were you given written information?

Question Three: Can you recall what type of observation level had been prescribed for you?

Question Four: Describe what happened during the observation?

  • Did the nurses talk with you?

  • Did you have an opportunity to talk about how you were feeling?

  • Could you leave the ward?

  • Could you attend planned activities, OT, or participate in group work with other service users?

  • Did you remain in your room or generally within the ward area?

Question Five: In your own words describe how you felt whilst on an observation level?

Question Six: What, if anything, could have made this experience better for you?



Adams, B (2000) Locked Doors or Sentinel Nurses? Psychiatric Bulletin 24, 327 _ 328

Altschul A (1972) Patient_nurse interaction: a study of interaction patterns in acute psychiatric wards. Edinburgh: Churchill Livingstone

Beck C B, Rawlins R P and Williams S R (1988) Mental Health _ Psychiatric Nursing. A holistic life-cycle approach. Missouri: CV Msoby Co.

Cormack, D (1976) Psychiatric Nursing Observed: a descriptive study of the work of the charge nurse in acute admission wards. London: Royal College of Nursing

CRAG Working Group on Mental Illness (1996), The Prevention and Management of Aggression _ a good practice statement. Scottish Executive Health Department, Edinburgh

CRAG/SCOTMEG Working Group on Mental Illness (1995), Nursing Observation of Acutely Ill Psychiatric Patients in Hospital _ A good practice statement. Scottish Executive Health Department, Edinburgh

Department of Health (2000) An Organisation with a Memory _ Report of an expert group on learning from adverse events in the NHS

Department of Health (2001) Building a Safer NHS for Patients

Dick, S and Keppie, M (1995) User Consultation Project _ Observation Policy _ Royal Edinburgh Hospital (unpublished)

Dodds, P and Bowles, N (2000) Dismantling Formal Observation and Refocusing Nursing Activity in Acute Inpatient Psychiatry: a case study. Journal of Psychiatric and Mental Health Nursing8, 183 _ 188

Duffy, D (1994) Report to the original good practice group

Duffy, D (1995) Out of the shadows: a study of the special observation of suicidal psychiatric in-patients. Journal of Advanced Nursing, 21 pp 944 _ 950

Fletcher, R F (1999) The process of constant observation: perspectives of staff and suicidal patients. Journal of Psychiatric and Mental Health Nursing, 6, 9 _ 14

Gillion, A (1999) Review of UK evaluative literature on clinical supervision in Nursing and Health Visiting, UKCC: London

Gournay, K and Bowers, L (2000) Suicide and self-harm in in-patient psychiatric units: a study of nursing issues in 31 cases. Journal of Advanced Nursing,32(1) 124 _ 131

Health Department (2000) Risk Management. Mental Health Reference Group, Scottish Executive

Health Department (2001) Caring for Scotland _ the Strategy for Nursing and Midwifery in Scotland

Jones, Jet al (unpublished) A study of psychiatric inpatients' experience of formal observation in acute inpatient settings _ Executive Summary

Learning Together _ A Strategy for Training and Lifelong Learning (1999)

Mental Welfare Commission (1992) Annual Report 1991/1992 Edinburgh HMSO

Mental Welfare Commission (1993) Annual Report 1992/1993 Edinburgh HMSO

Mental Welfare Commission for Scotland (2001) General Hospitals and the Mental Health (Scotland) Act, 1984 Guidance circulated to Chief Executives of Trusts

Mental Welfare Commission for Scotland (1999) Locked Doors in Hospital _ Survey. Annual Report, Visiting Programme 1998 _ 1999

National Confidential Inquiry into Suicide and Homicide by People with Mental Illness _ Review of Progress 1996 _ 2000

National Confidential Inquiry into Suicide and Homicide by People with Mental Illness Safety First (2001)

Neilson, P and Brennan, B (2001) The use of special observations: an audit within a psychiatric unit, Journal of Psychiatric and Mental Health Nursing, 8, 147 _ 155

Peplau, H (1988) Interpersonal relations in nursing 3rd edition. London: Macmillan

Pitula C R and Cardell, R (1996) Suicidal Inpatients' Experience of Constant Observation, Psychiatric Services Vol. 47, No 6, 649 _ 651

Porter, S, McCann, I. and McGregor Kettles, A (1998) Auditing Suicide Observation Procedures. Psychiatric Care, 5 (1) 17 _ 21

Reynolds B (1985) Issues arising from teaching interpersonal skills in psychiatric care for the suicidal patient. Journal of Advanced Nursing, Chapter 13 of Kagan C (ed)

Scottish Health Advisory Service (1999) Mental Health Standards _ Edinburgh

Scottish Health Advisory Service (2001) Quality Indicators and Self Assessment Framework. Edinburgh

Standing Nursing and Midwifery Advisory Committee (1999) Mental Health Nursing: "Addressing Acute Concerns" Report by the Department of Health




The essentials of this report are as follows.

There are four general principles:

  • information from a single source is never going to be enough to assess risk

  • and corroboration will always have to be sought;

  • similarly one person alone cannot perform an adequate risk assessment

  • and access to the network of people surrounding an individual is crucial;

  • people who present a risk to others are also likely to be vulnerable to self-harm

  • self-neglect or exploitation; (in other words the perception of others should not be allowed to blot out the possibility of that individual also needing protection);

  • factors such as age

  • gender and ethnicity are unreliable predictors of risk to harm to others.

In the history taken from an individual being assessed, certain items must be sought:

  • previous violence or suicidal behaviour;

  • "social restlessness" _ few relationships, frequent changes of address or employment;

  • evidence of poor engagement with mental health services;

  • presence of substance misuse;

  • a social background promoting violence;

  • any precipitants or changes in mental state or behaviour that have occurred prior to previous episodes of violence or relapse;

  • recent change in any of these risk factors;

  • evidence of recent severe stress, especially major losses;

  • evidence that medication has recently been discontinued.

It is important to identify potential victims, particularly those who figure in abnormalities in the patient's mental state (for example, the focus of delusions or the apparent source of hallucinations). In the patient's mental state the emotionality with which the patient presents (for example irritability, anger, hostility or suspicion) is important, as are specific threats made by the patient. Also, beliefs of threat, or persecution or control of mind or body by external forces are noteworthy.

In recording the assessment the following points have to be noted:

  • how serious is the risk?

  • is the risk specific to one person or situation, or is it general?

  • how immediate is the risk?

  • how volatile is the risk?

  • what potential factors increase the risk, and what might decrease it?

  • what specific treatment, and which management plan can best reduce the risk?

In managing risk, there are two basic principles:

  • a person working within a mental health service

  • having identified the risk of dangerous behaviour

  • has a responsibility to take action with a view to reducing that risk and managing it effectively; and

  • in managing risk

  • the emphasis should be towards safety. That starts by engendering a relationship with the patient which makes him or her feel safer and less distressed.

Considerations for managing risk include:

  • does he or she require admission as an inpatient?

  • should he or she be detained under the Mental Health (Scotland) Act 1984?

  • what level of physical security is likely to be needed?

  • what level of observation is required?

  • what medication should be used?

  • it should be understood clearly by ward staff how the medication is to be employed;

  • if there is another episode of violence, how should it be managed?

If the patient is being managed in the community, other questions come to the fore:

  • is there a place for the Care Programme Approach?

  • can the Mental Health (Scotland) Act be used, or is there a case for a community care order?

  • what community supports are available, how effective might they be, and how can they best be assisted?

  • do the carers and family have access to appropriate support and help?

  • have the carers _ in the family, and in other agencies _ been adequately informed about the situation, how it is likely to develop, and what help they can expect to receive?

Fundamental to the management of any situation is:

  • the plan of management clearly recorded in an accessible place, in legible writing;

  • the date for review of the assessment and management plan should be set down, after agreement with all those involved. That date needs to be passed on to all those who need to know;

  • the patient's general practitioner must be informed;

  • individuals who should or are entitled to receive information should be identified and responsibility assigned to carry this out;

  • the threshold for breaching confidence to ensure public safety has been defined;

  • if responsibility for the management of a plan of action is being passed on to another team or individual, it must be accepted explicitly. The information passed on must include all relevant details.


The Clinician

  • to respond as rapidly as possible when concern is expressed by a colleague or member of staff from a partner agency about an increased risk from a patient;

  • always to make a systematic assessment;

  • always to consult as widely as is possible and appropriate;

  • not only to make a decision on what needs to be done, but to make explicit the reasons for that decision and to write them down;

  • to make a management plan based on the assessment;

  • to record details of the management plan;

  • to share the management plan as appropriate with all those who have a legitimate concern with its implementation;

  • to make no assumptions about what other people will do _ if their co-operation is required in carrying out a management plan, make sure that there is explicit consent;

  • to make an appropriate arrangement for monitoring the management plan, making sure that a date is set and kept for subsequent review.

Clinical Teams

  • should have an agreed protocol for responding to patients showing significant risk. This protocol should identify:

    _ the appropriate senior clinicians to be contacted to conduct assessment or re-assessment;

    _ the means by which they should be contacted must be clear;

    _ if the identified person is not contactable, a subsidiary route should be available;

    _ to have agreed protocols for follow-up and review of patients;

    _ to establish and maintain links with other agencies, based on mutual respect for the contribution which can be made, to involve them in the care and management of patients who present a significant risk.

Service Managers

  • the effective assessment and management of people presenting increased risk of harm should be of the highest priority for allocation of resources;

  • risk assessment and clinical risk management is time-consuming and expensive. The appropriate resources should be made available;

  • proper assessment and management of clinical risk cannot take place in an unsafe environment or within inadequate facilities;

  • senior staff must be expected always to be available to take responsibility for decisions about assessment and management of risk;

  • training must be supported and adequately resourced;

  • alliances and partnerships with other agencies should be maintained, and mechanisms put in place to ensure their maintenance.



(from "New Directions": Report on the Review of the Mental Health (Scotland) Act* 1984)

1. Non-discrimination

People with mental disorder should, whenever possible, retain the same rights and entitlements as those with other health needs.

2. Equality

All powers under the Act* should be exercised without any direct or indirect discrimination on the grounds of physical disability, age, gender, sexual orientation, race, colour, language, religion or national or ethnic or social origin.

3. Respect for Diversity

Service users should receive care, treatment and support in a manner that accords respect for their individual qualities, abilities and diverse backgrounds and properly takes into account their age, gender, sexual orientation, ethnic group and social, cultural and religious background.

4. Reciprocity

Where society imposes an obligation on an individual to comply with a programme of treatment and care, it should impose a parallel obligation on the health and social care authorities to provide appropriate services, including ongoing care following discharge from compulsion.

5. Informal Care

Wherever possible care, treatment and support should be provided to people with mental disorder without recourse to compulsion.

6. Participation

Service users should be fully involved, to the extent permitted by their individual capacity, in all aspects of their assessment, care, treatment and support. Account should be taken of their past and present wishes, so far as they can be ascertained. Service users should be provided with all the information necessary to enable them to participate fully. All such information should be provided in a way which renders it most likely to be understood.

7. Respect for Carers

Those who provide care to service users on an informal basis should receive respect for their role and experience, receive appropriate information and advice, and have their views and needs taken into account.

8. Least Restrictive Alternative

Service users should be provided with any necessary care, treatment and support both in the least invasive manner and in the least restrictive manner and environment compatible with the delivery of safe and effective care, taking account where appropriate of the safety of others.

9. Benefit

Any intervention under the Act* should be likely to produce for the service user a benefit which cannot reasonably be achieved other than by the intervention.

10. Child Welfare

The welfare of a child with mental disorder should be paramount in any interventions imposed on the child under the Act *.

* The Act refers to the new Act, which will arise from the report, not the existing 1984 Act

Back to top