Article Text

Download PDFPDF

Child protection medical assessments: why do we do them?
  1. Charlotte B Kirk1,
  2. Angela Lucas-Herald2,
  3. Jacqueline Mok1
  1. 1Department of Community Child Health, Royal Hospital for Sick Children, Edinburgh, UK
  2. 2Faculty of Medicine, University of Edinburgh, UK
  1. Correspondence to Dr Charlotte B Kirk, Department of Community Child Health, Royal Hospital for Sick Children, 10 Chalmers Crescent, Edinburgh EH9 1TS, UK; charlotte.kirk{at}


Introduction Child protection guidelines highlight the importance of medical assessments for children suspected of having been abused.

Aim To identify how medical assessments might contribute to a diagnosis of child abuse and to the immediate outcome for the child.

Method Review of all notes pertaining to medical assessments between January 2002 and March 2006.

Results There were 4549 child protection referrals during this period, of which 848 (19%) proceeded to a medical examination. 742 (88%) case notes were reviewed. Of the medical examinations, 383 (52%) were for alleged physical abuse, 267 (36%) for sexual abuse and 20 (3%) for neglect. 258 (67%) of the physical abuse cases were considered to have diagnostic or supportive findings as compared to 61 (23%) of the sexual abuse cases (χ2=146.31, p<0.001). In diagnostic or supportive examinations or where other potentially abusive concerns were identified, 366 (73%) proceeded to further multi-agency investigation and 190 (41%) to case conference. 131 (69%) of these resulted in the registration of the child on the child protection register. Other health concerns were identified in 121 (31%) of physical and 168 (63%) of sexual abuse cases.

Conclusion In this case series, 465 (63%) out of 742 examinations showed signs diagnostic or supportive of alleged abuse or highlighted other abusive concerns. This endorses the view that medical examination is an important component in the assessment of child abuse as it provides information to support or refute an allegation and helps to identify the health and welfare needs of vulnerable children.

Statistics from

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.


A comprehensive medical assessment is an essential component of the multi-agency investigation of child abuse. The purposes of this assessment are to establish the need for immediate treatment and secure ongoing medical care, to provide information to support a diagnosis of child abuse and evidence to sustain criminal proceedings, and to provide reassurance to the child and their family.1 A medical examination for allegations of abuse, particularly sexual abuse, may be seen by some professionals as invasive or even harmful to the child. However, it often reassures that no long-term physical damage or health risk has occurred and when conducted sensitively may be the start of a healing experience for both the child and their family.2 The medical assessment should also aim to identify health and welfare needs in a group of very vulnerable children and is integral to the child protection process.

What is already known on this topic

  • Child protection guidelines recommend that medical assessment is included in the investigation of alleged abuse.

  • The majority of medical examinations for alleged sexual abuse have non-specific findings.

  • Children examined for alleged abuse may have unrecognised health problems.

What this study adds

  • Two thirds of medical examinations for alleged physical abuse demonstrate supportive findings.

  • Just under half of all children examined had unmet health or welfare needs.

  • Investigations with positive findings at medical examination are more likely to go to child protection case conference.

Almost all studies of child protection medical assessments have focused on sexual abuse inquiries, how often physical signs are identified and whether this increases the likelihood of a perpetrator conviction. The results are conflicting. De Jong and Rose concluded that a child's testimonial was the most important evidence for obtaining a conviction, as physical signs were present in only 23% of cases where a conviction was achieved.3 However, Palusci et al showed that positive clinical findings were 2.5 times more likely to result in a criminal prosecution with a finding of perpetrator guilt.4 Heger et al found 96.3% of children referred for evaluation had a normal examination, which led the authors to conclude that the child's history remained the most important diagnostic factor.5

Girardet and co-workers identified other benefits of a medical assessment for alleged sexual abuse.6 A medical or psychological diagnosis judged by the examiner to require intervention was made in a quarter of children seen. By contrast, only 9% of children had probable or definite physical or laboratory evidence that supported the allegation of sexual assault. The authors argued that unmet healthcare needs identified at examination of these children were as valid a reason for performing the examination as identifying physical findings.

These studies were conducted in the USA where different healthcare and legal systems might have influenced the results. Only one study has looked at this subject for physical abuse cases and this concerned the variability of medical opinion.7 To investigate local practice in both physical and sexual abuse cases, we performed a retrospective case note review of all medical assessments following a child protection referral over a 51-month period. We aimed to evaluate the nature and frequency of positive clinical findings and whether these influenced the outcome for the child, and to identify the unmet health and welfare needs of our population.


In Edinburgh all child protection referrals proceed through an inter-agency referral discussion (IRD). This is a tri-partite documented discussion between social work, police and health where information is shared and a course of action decided as soon as is practically possible.8 If the consensus decision is that a medical assessment is required, all children are examined either by a paediatrician alone or with a forensic medical examiner. All examination findings are recorded using a standardised clinical proforma. Basic demographic data about all IRDs are logged in a departmental Microsoft access database for clinical use. A list was compiled of all medical examinations generated by child protection referrals in the Edinburgh area between 1 January 2002 and 31 March 2006 and case notes obtained.

A data collection form was designed to record basic demographic characteristics including the type of alleged abuse. In the absence of any standards, clinicians tended to document their opinions using descriptive terminology and ordinal scales. The conclusion of the medical report was used to classify examination findings into subcategories of sexual and physical abuse (table 1).

Table 1

Categories of abuse

Occasionally, although examination findings were not consistent with the allegations made, other signs were found that were likely to be due to abuse (eg, signs of neglect in a child with a physical injury). We also recorded if the examination was abandoned due to lack of consent from the young person. Any other medical concerns identified at examination were categorised as follows: behavioural problems, poor growth, medical concerns (including possible sexually transmitted disease), developmental delay or poor dental hygiene. Any medical follow-up initiated by the examination was documented.

The immediate outcome for the child after the examination was also recorded. Actions included further police and social work investigation, hospital admission for treatment or for the child's protection, or if the child was placed in alternative care. In some cases, no further action was taken. Medium-term outcomes documented were whether investigations proceeded to case conference and the decision made at case conference.

The information was extracted and recorded by two of the three investigators in a secure Microsoft access database. Difficulties in interpreting the notes were resolved by consensus discussion. The study was lodged and approved as an audit project with the clinical governance support team for NHS Lothian and was deemed not to require NHS ethics review by the South East Scotland Research Ethics Service. Data were anonymised and managed in accordance with local data protection procedures as defined by the Caldicott guardian. Association between factors was tested by t test or χ2 tests as appropriate.


Between 1 January 2002 and 31 March 2006, 4549 child protection referrals were received from a population base of approximately 130 000 children. Of these, 848 (19%) proceeded to a medical examination. In the remaining cases tripartite discussion deemed a medical assessment would not contribute to the investigation or be in the child's best interest. We obtained and reviewed the notes of 742 (88%) children. Excluded were 106 children whose case notes could not be traced.

Of the 742 included cases, 440 (59%) were girls and 302 (41%) were boys. The mean age of the boys was 5.3 years (SD 4.2) and of the girls 8.6 years (SD 5.4). A two-sample t test demonstrated that this is a highly significant age difference (t=8.72, p<0.001).

The majority of referrals were for physical and sexual abuse, 383 (52%) and 267 (36%), respectively. Other referrals were for a combination of physical and sexual abuse (8.3%), neglect (3%) and for those girls aged >13 years where consenting sexual intercourse had taken placed but there were concerns that the relationship was potentially abusive (1%). Referrals of girls aged >10 years were mainly for sexual abuse and for those aged 0–5 years of both sexes, alleged physical abuse was the commonest cause of referral. Assessments were joint paediatric and forensic medicals in 509 cases and single doctor medicals in 233. No children were examined only by a forensic medical examiner.

Results of medical assessments are shown in table 2. Of 383 children examined for physical abuse, 67% had findings that were either supportive or diagnostic of the alleged abuse. Those injuries described as ‘no evidence of alleged abuse’, were either typical of normal play or there was a witnessed and/or credible parental story of injury. In 267 children examined for alleged sexual abuse, the majority were felt to have findings which were non-specific (genital findings were normal in children who had given a detailed disclosure of sexual abuse). Only two children who disclosed sexual abuse had diagnostic findings and a quarter had findings that were described as supportive of the alleged abuse. The difference in positive findings between those examined for physical (258/383) and sexual abuse (61/267) was extremely significant (χ2=146.31, p<0.001).

Table 2

Results of medical assessments

Figure 1 shows that 338 (46%) of the 742 children examined had another health need identified as a result of the medical assessment. Of the 168 (63%) children examined for alleged sexual abuse, 126 (47%) had unmet medical needs. This included children requiring follow-up through genito-urinary services for sexual health counselling and screening for sexually transmitted disease. A total of 615 children (83%) were referred for follow-up as a direct result of the medical assessment. The majority (453, 61%) were seen by the community paediatric services, 61 (8%) were referred for therapeutic support by child and adolescent mental health and 101 (14%) were referred to hospital clinics including services for adolescent sexual healthcare.

Figure 1

Health needs identified at medical assessment.

The immediate outcome for children following medical assessment is shown in table 3 and figure 2. Irrespective of clinical findings, further multi-agency investigation occurred in three quarters of the referrals. Single agency follow-up only (usually by social work) occurred in a very small proportion of those whose findings were supportive of abuse and none of those whose findings were diagnostic of abuse. Immediate outcome was similar across different categories of abuse, although a greater number of physical abuse children were admitted or accommodated (70/383 (18%) compared to 9/267 (3.3%) in those assessed for sexual abuse). No information about follow-up was available in 19 children who were from outside the Edinburgh area.

Figure 2

Immediate outcome according to type of abuse.

Table 3

Immediate outcome for children by examination finding

Table 4 shows that 41% of all investigations proceeded to child protection case conference. This was significantly more likely to occur if the child had supportive or diagnostic findings at examination or other concerns of abuse identified. In these cases, 190/465 proceeded to case conference as compared to 39/277 of cases where findings were non-specific or not consistent with abuse (χ2=17.54, p<0.001). At case conference a greater percentage of those with positive findings were placed on the child protection register, although this difference did not reach statistical significance.

Table 4

Child protection procedures by clinical findings


In Edinburgh, child protection procedures dictate that when there is concern about a child, inter-agency discussions must occur between professionals in health, social services and police to enable a joint decision about the need for medical assessment. Strict adherence to the procedures and good inter-agency working have led to complete ascertainment of children who are likely to require medical examination following an allegation of abuse. This is the first study, to our knowledge, to examine the frequency of positive findings and the outcome of child protection medical examinations in children who allege physical abuse. We found that over two thirds of children seen for allegations of physical abuse had clinical findings that were described as supportive or diagnostic of abuse. While the majority of medical assessments for physical abuse provide important information to sustain the diagnosis of abuse, positive examination findings were identified in only one fifth of children who alleged sexual abuse. Another third had clinical signs which were non-specific, the usual phrase used when the findings ‘could neither confirm nor refute the allegation’.

This is in keeping with results from other studies. It is widely accepted that a clear statement from the child remains the single most important diagnostic feature in sexual abuse. In a 5-year prospective study of 2384 children referred for evaluation of possible sexual abuse, 95.6% had normal examinations.5 Blinded review of examination photographs of children with a perpetrator conviction for sexual abuse reported that 28% of photographs were normal and 49% had non-specific changes only.9 Children who are seen within 7 days of the alleged incident are more likely to have positive findings10 and therefore best practice must ensure medical examination occurs as soon as possible after the alleged incident.

Where there were concerns about neglect, medical examinations were conducted in only 20 children. The small numbers prevent meaningful comment about this group, but supportive findings or other concerns were present in almost two thirds, underlining the importance of a comprehensive medical assessment in this group of vulnerable children. To our knowledge there are no published studies which examine the outcome of medical examinations for children who suffer alleged neglect.

Our results support the view that medical assessment has a valuable role in the child protection procedures beyond the forensic requirement. Health problems were identified in a group of children who are likely to have poor access to medical care. Sixty two per cent of children referred for alleged sexual abuse and 31% referred for physical abuse had a previously unrecognised health need. The majority were described as unmet medical needs, although these included children referred for investigation of sexually transmitted disease. Over 80% of children had medical or psychological follow-up instigated as a direct result of the medical assessment. Similar findings were presented in a review of 473 children examined for alleged sexual assault. A medical or psychological diagnosis that required intervention from the examiner was made in 123 children (26%). Thirty nine children had a diagnosis that could result in significant morbidity if not immediately addressed.6 As in our population referred with alleged sexual assault, a greater percentage had unmet health needs than clinical findings diagnostic or supportive of abuse. The function of a paediatrician at the medical examination is to identify these needs, organise follow-up and in so doing, provide an excellent opportunity for health surveillance.

Obtaining information about the outcome of child protection investigations is difficult. We have shown that one fifth of children with findings supportive or diagnostic of abuse were immediately accommodated or admitted to hospital after the examination and three quarters had continuing police and/or social work investigation. This included a small number of children who required admission for treatment after severe physical abuse. Some children who had diagnostic findings were not accommodated immediately because the alleged perpetrator was removed from the household. In contrast, a third of those with findings which did not indicate abuse had no further action taken after examination. Positive examination findings (ie, diagnostic, supportive or other signs identified) also resulted in significantly more child protection conferences taking place. Outcome at case conference, however, was not related to whether there were positive findings at examination. The fact that results of medical examination do not seem to directly influence the outcome emphasises the role of the medical contribution as only one piece of the diagnostic jigsaw in child protection.11

Our study has some limitations. Concluding remarks made by the medical examiners were used to categorise findings at examination. This relied on the correct interpretation of signs by those examining the children and we did not draw our own objective conclusions from the findings. There are proposed classification systems for the assessment of sexual abuse but not physical abuse.12 In the absence of standards, Lindberg et al showed that in cases of physical abuse, even experienced clinicians showed wide variability in interpretation of terms such as ‘definite abuse’ or ‘reasonable concern for abuse’.7 A prospective study using a standardised data collection form might address these issues and result in a more robust study.

We obtained information about the immediate outcome for children and whether investigations proceeded to case conference only from the medical case notes. Our information would be more complete if we also used social work and police records. This would require the co-operation of police, social work and the legal profession, but might help identify whether positive clinical findings influence the decision to prosecute the case.


We have shown that 63% of medical assessments conducted for child protection purposes result in signs which support a diagnosis of child abuse and 45.5% of children have new health needs identified at this assessment. In all types of alleged abuse supportive signs increase the likelihood of case conference. Professionals in social work, police and the legal profession place considerable emphasis on the significance of clinical findings and tend to focus on the evidential needs of the medical examination. In addition, carefully conducted medical assessments provide an important function in identifying unmet health needs and undiagnosed medical problems in a group of very vulnerable children. The importance of having paediatric expertise at the examination is highlighted and this could have resource and training implications.


The authors thank the Child Protection administrative staff, especially Lorraine Johnston, for their time in seeking out the case notes of children in the study.



  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.