Performance of screening tests for child physical abuse in accident and emergency departments

Authors: Woodman J, Pitt M, Wentz R, Taylor B, Hodes D, Gilbert RE

Journal: Health Technology Assessment Volume: 12 Issue: 33

Publication date: December 2008



Woodman J, Pitt M, Wentz R, Taylor B, Hodes D, Gilbert RE.Performance of screening tests for child physical abuse in accident and emergency departments. Health Technol Assess 2008;12(33)

Download: Citation (for this publication as a .ris file) (3.4 KB)

Journal issues* can be purchased by completing the form.

The cost of reports varies according to number of pages and postage address. The minimum cost for a copy sent to a UK address is £30.00. We will contact you on receipt of your completed form to advise you of actual cost. If you have any queries, please contact

*We regret that unfortunately we are unable to supply bound print copies of Health Technology Assessment published before issue 12:31. However, PDFs are available to print from the "Downloads" tab of the issue page.


No responses have been published. If you would like to submit a response to this publication, please do so using the form below.

Comments submitted to the NIHR Journals Library are electronic letters to the editor. They enable our readers to debate issues raised in research reports published in the Journals Library. We aim to post within 2 working days all responses that contribute substantially to the topic investigated, as determined by the Editors.

Your name and affiliations will be published with your comment.

Once published, you will not have the right to remove or edit your response. The Editors may add, remove, or edit comments at their absolute discretion.

Post your response



Middle Initial

Occupation / Job title

Affiliation / Employer



Other authors

For example, if you are responding as a team or group. Please ensure you include full names and separate these using commas

Statement of competing interests

We believe that readers should be aware of any competing interests (conflicts of interest).

The International Committee of Medical Journal Editors (ICMJE) define competing interests as including: financial relationships with industry (for example through employment, consultancies, stock, ownership, honoraria, and expert testimony), either directly or through immediate family; personal relationships; academic competition; and intellectual passion.

If yes, provide details below:

Enter response title

Enter response message


Security key

Regenerate security key

By submitting your response, you are stating that you agree to the terms & conditions

The full text of this issue is available as a PDF document from the Downloads section on this page.



To determine the clinical effectiveness of screening tests for physical abuse in children attending accident and emergency (A&E) departments in the UK.

Data sources

Searches were limited to studies published after 1974 and were carried out from August 2004 to October 2006 using the following methods: searching electronic databases, searching the publications catalogue of the NSPCC, scanning reference lists, hand-searching journals, searching the internet, approaching professional contacts for unpublished data, and searching in three key journals.

Review methods

A simple decision-analytic model was used to integrate the findings of nine systematic reviews regarding the incidence of physical abuse, the characteristics of children attending A&E, and the performance of screening tests for physical abuse.


A total of 66 studies, including 11 unpublished studies, were included in the nine systematic reviews. Overall the quality was poor. There was consistent evidence that physical abuse affects about 1 in 11 children in the UK each year. The proportion of abused children requiring medical attention is small but poorly quantified. Approximately 1% of all attendances of injured children at A&E are for physical abuse. There was clear evidence that physically abused children attending A&E are missed, but the performance of the clinical screening assessment was poorly quantified. There was no evidence that any test was highly predictive of physical abuse. Among severely injured children admitted to hospital, those under 1 year were more likely to be abused than older children. However, evidence that young age was a risk factor for abuse among all injured children attending A&E was inconsistent. There was weak evidence that a community liaison nurse improved the performance of the screening assessment in A&E, and it was estimated that combining a nurse with the standard screen would result in referral to social services of about half of the abused children attending A&E. However, given the poor quality of the data, this is highly uncertain. The addition of screening protocols to the clinical screening assessment offered marginal benefits, and additional false-positive referrals exceeded additional abused children detected. The benefits of protocols declined as the accuracy of the clinical screening assessment improved. The most effective protocol was to refer all injured infants and children who were social work active.


Improving clinical screening assessment is likely to be more useful than protocols in improving the detection of physically abused children attending A&E. Further improvements might be achieved by following up children referred to paediatricians for suspected abuse who fail to reach the high level of certainty required to justify referral to social services. Many professionals voiced a need for access to experienced social services advice that is not under pressure to minimise referrals to an overloaded service, and consideration might be given to making such advice centrally available.

Share this page

Email this page
Publication updates

If you would like to receive information on publications and the latest news, click below to sign up.