Article Text

Download PDFPDF

Patient self discharge from the emergency department: who is at risk?
  1. V L Henson1,
  2. D S Vickery2
  1. 1Gloucestershire Royal Hospital, Intensive Care, Gloucestershire, UK
  2. 2Queens Medical Centre, Nottingham, UK
  1. Correspondence to:
 Dr V Henson
 Emergency Department, Cheltenham General Hospital, Sandford Road, Cheltenham, Gloucestershire GL53 7AN;


Objectives: To identify the number of patients who take their own discharge, to review how their competence1,2 to make the decision to refuse treatment was being assessed and managed, and to quantify the medical risk to these patients.

Methods: A retrospective review of all case notes coded as “self discharge” for a three month period. A second cohort was reviewed following the introduction of a new self discharge proforma.

Results: Patients who self discharge represented 0.5% of the total number of attendances, and those who “did not wait” represented a further 11%. Fifty percent were under the influence of drugs or alcohol. Sixty three percent refused admission, 24% refused treatment, and 13% refused medical assessment. In the first audit, assessment of capacity was documented in 0%, 58% signed a self discharge form, 25% were reviewed by a doctor prior to leaving, and 31% left the department without the staff’s knowledge. In the second audit, capacity was documented in 80%, 80% signed the new self discharge form, 41% were reviewed by a doctor prior to leaving, and 0% left the department without being reviewed.

Discussion: Self discharge may be a significant risk to both the patient and the hospital trust. Departments should review their own systems for assessing and managing this patient group. Further work looking at the consequences of self discharge is required.

  • capacity
  • consent
  • risk management
  • self discharge

Statistics from

A significant proportion of patients who present to the emergency department take their own discharge against the advice of nursing and medical staff. Of particular concern are those patients who refuse treatment or hospital admission. Although patients are under no obligation to follow medical advice, it is crucial that they are able to understand the implications of this decision. In other words, are they competent to refuse treatment and/or admission?

We performed an audit to determine the extent and nature of cases that self discharged from a District General Hospital. Following analysis of the first stage of the audit a self discharge proforma was introduced, which included advice on the assessment of competence.3,4


Over a three month period all attendances to the emergency department coded as self discharge were audited. Outcome measures were presenting complaint, triage category, extent of medical or nursing assessment, and capacity to refuse treatment or admission.

Following data analysis, a self discharge proforma (fig 1) was implemented and the case notes reviewed for a subsequent three month period.

Figure 1

 Example of the new self discharge proforma


There were 55 cases in the first audit and 56 in the second, which represented 0.5% of all attendances during these periods. Patient demographics were very similar.

Sixty per cent of the patients were male, no patient was under the age of 16 years, and 7% were aged 56 years or over. Fifty nine per cent attended between 8 pm and 8 am. Fifty per cent of all cases were documented to be under the influence of drugs or alcohol.

Hospital admission for inpatient care was refused by 63% of patients. In addition, 24% of patients refused treatment (appropriate wound closure, dressings, antibiotics, etc). The remaining 13% of patients did not wait for medical assessment but had been assessed by the nursing staff and initial investigations commenced—for example, baseline observations, urinalysis, and electrocardiogram.

Manchester triage category

Analysis of cases according to their triage category is shown in table 1.

Table 1

 Analysis of cases according to their triage category

Presenting complaint

Distribution of cases according to the presenting complaint is shown in table 2. The category “medically unwell” includes conditions such as asthma, seizures, and stroke.

Table 2

 Distribution of cases according to the presenting complaint

Assessment and documentation

In the first audit, 58% of patients signed the self discharge form (fig 2) and 31% left the department without the staff’s knowledge—that is, nothing written in the notes to say the patient had self discharged, simply coded as self discharge with no further documentation of ongoing management. In the second audit, 80% of the new self discharge forms were completed and no patient left the department without the staff’s knowledge—that is, all notes documented patient left the department against medical advice, whether or not the patient was reassessed, and if they signed or refused to sign the self discharge form. In the second audit, 41% of patients were reviewed by a doctor prior to leaving compared with 25% in the first audit.

Figure 2

 Example of the self discharge form used in the first audit

Capacity to refuse treatment or admission

In the first audit, none of the notes made any reference to the patients’ capacity to refuse consent but in the second audit capacity was formally documented in 82% of cases. None of these cases were deemed incapable of refusing consent.


There may be a significant risk associated with the self discharge of an emergency patient. Not only to a sick or injured individual, but also to the hospital trust if an adverse clinical event occurs. Patients may not be in an appropriate state of health or mind and the emergency department staff may not have a robust system for assessing the capacity of the patient to refuse treatment and admission. Often these cases occur late in the evening or overnight when more junior and less experienced doctors may be involved.5,6

In this audit, 67% of patients presented with a medical illness, chest pain, or following an overdose/deliberate self harm. One pertinent clinical case was of a 40 year old male with new onset of chest pain who self discharged, only to present one week later with an acute myocardial infarct. The preponderance of “medical” cases that refused treatment and admission is striking. Fifty per cent of all the cases were not under the influence of drugs or alcohol, and a minority presented following an assault or minor trauma. Thus, it does not appear to be the stereotypical drunk or drug taking patient who has fallen or been in a fight that refuses medical attention.

With 63% of the self discharge patients refusing hospital admission, the risk of an adverse clinical outcome is appreciable. Thirty three per cent of these patients were assessed as either triage category 1 or 2. It is incumbent on emergency department staff to ensure that patients understand the treatment or care that is advised and the potential consequences of not receiving it. The first stage of the audit showed that in 31% of cases the patient simply left and the opportunity to review the risk was lost. In those that expressed the intention to discharge themselves, we suspect that 58% of the patients simply signed the standard self discharge form as a cursory matter with minimal assessment of their capacity to safely do so.

The introduction of a much more comprehensive form heightened awareness with the result that nobody left without staff knowledge. In 80% of cases capacity, an explicit indication of medical risk and prompts to try to minimise this risk were utilised. One of the authors carried out a telephone survey of emergency departments in the south west region asking for a faxed copy of their standard self discharge documentation. Of the eight departments that responded, none of the forms included assessment or reference to the patient’s capacity to refuse consent, or the potential seriousness of their medical complaint. Only one trust included discharge advice and follow up arrangements.

If it is recognised that the standard self discharge forms used in many emergency departments across the country have no useful clinical role, and have little legal weight in protecting the trust, we question why they are used at all. Emergency departments should change their practice of assessing and documenting the capacity of these patients, as recommended by the defence agencies,7,8 the Law Society, and the BMA,9 and review their paperwork to help staff manage these patients more appropriately.



  • Competing interests: none declared

Request Permissions

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

Linked Articles

  • Primary Survey
    Kevin Mackway-Jones Geoff Hughes
  • Editorial
    G Hughes K Mackway-Jones