• Open Access

A pilot study to measure marks in children with cerebral palsy using a novel measurement template



Therese Bennett, Children's Community Services, Central Manchester University Hospitals NHS Foundation Trust, Newton Heath Health Centre, 2 Old Church Street, Newton Heath, Manchester M40 2JF, UK

E-mail: therese.bennett@cmft.nhs.uk



The primary aim of this pilot study was to trial a method of assessing bruises in a population of disabled children. If the method was found to be sufficiently robust it would be our intention to undertaking a more extensive observational study.


Less is known about normal bruising patterns in children with disability than in those without. It is important that the method used to assess bruising is objective and repeatable. In an effort to define and improve repeatability, we employed a novel bruise measurement template which was printed onto transparent acetate sheets.


Twenty primary school age children, the majority of whom were non-ambulant and severely disabled with cerebral palsy, underwent full skin examination. The template was used to assess any bruises seen. A comparison was then made between measurements made by experienced paediatricians using the template and using a standard tape measure on a series of bruise images in 25 photographs.


The majority of children in our pilot were found to have bruises, with one child having 6 and one 7 bruises. This comparative study showed that the two techniques had a very similar precision and that the template was easy to use. Greater precision would require a tighter measurement protocol, whether with a template or a tape measure.


Further evaluation of the application of such a template would be worthwhile. We would suggest that our finding of some bruising in this population of disabled children is borne in mind whenever bruising is found in a non-ambulant child.


The systematic review by Maguire and colleagues (2005) of bruising in non-disabled children provides an evidence base to help differentiate abusive from non-abusive bruising. There is some information about bruising in disabled children with a small case series by Barton and Finlay (2005) suggesting that ‘bruising in children with disability or special needs may be very different from bruising patterns in their peer group’ and a recent study in children with motor disabilities by Newman and colleagues (2010) finding ‘a progressive reduction in the number of skin injuries with decreasing mobility’. It is also generally accepted that children with disability are more likely to suffer abuse than their non-disabled peers: NSPCC (2003) and Sullivan and Knutson (2000) reported that physical abuse was 3.8 times more likely. Maguire and colleagues (2005) found that ‘bruises were characteristically small in non-abused children’ and that ‘abusive bruises were often larger’ and Barber and Sibert (2000) stated that ‘careful examination of the child, documentation of bruising with use of body charts depicting size in two planes’ is needed.

With a view to undertaking a more extensive observational study of these issues, we have carried out a modest pilot study of bruising in 20 children with cerebral palsy (CP). The primary aim was to test a novel measurement template.

If observations of bruises on different children at different times by different examiners are to be compared, then it is important that an objectively repeatable protocol be employed. The default method is to use a tape measure, but repeatability is inherently likely to be limited as bruises come in all shapes and sizes and they have poorly defined edges. In an effort to improve, or at least define, repeatability, we devised a measurement template which was printed onto transparent acetate sheets. The precision of this technique and the default technique using a tape measure was subsequently checked against a photographic sample of bruises.


Part 1

The pilot study took place between January and December 2007 in two specialist primary schools in Manchester. There were 36 children with CP attending these schools at the time.

All children recruited had a full skin examination and the location of any bruise or mark was recorded on a skin map and measured using the template. Information about the child's disability was recorded and their motor skills development was assessed using the Gross Motor Function Classification System (GMFCS) as described in Russell and colleagues (2002). Children at Levels 1 and 2 of the GMFCS are ambulant, with Level 2 children requiring assistive devices, for example ankle-foot orthoses (AFOs), to enable them to walk.

Parents were informed about the study at the time of their child's scheduled paediatric review in school with one of the four examiners and written consent obtained. Consent also included permission for the first author to make contact by telephone afterwards to seek parents’ opinion about the study.

The children were examined by one of four paediatricians (the examiners) experienced in both child protection medical examinations and the medical management for children with complex disability.

The template was a series of 12 circles graded in diameter from 5 mm to 55 mm printed onto a transparent A4 acetate sheet (Fig. 1). An additional sheet was available for larger bruises if necessary but was not required. An increment of 5 mm was chosen so that a useful range of bruises could be measured with a single template; as will be seen, this was not a limiting factor in the ultimate precision of the measurement. The sheet was overlaid on any bruise or mark seen and two measurements were recorded: the size of the largest circle that could be fitted wholly within the bruise (minimum bruise diameter) and the smallest circle that could wholly enclose the bruise (maximum bruise diameter).

Figure 1.

Bruise-measuring template.

This part of the study was approved by the local ethics committee and was supported by school staff and governors.

Part 2

The consistency of measurements between different users was then tested both for the template and for a standard tape measure. For this part of the study, a series of approximately life-size colour photographs of 25 bruises taken of children who had attended the local child protection clinic in recent years was prepared. The photographs were taken for clinical or forensic reasons; consent included use for research purposes. A set of photographs was given to 19 paediatricians. The doctors were asked to record the minimum and maximum diameter of each bruise using the template. This exercise was later repeated with the same doctors being asked to estimate the length and width of each bruise using a tape measure graduated in mm. The second exercise took place 9 months after the first, this being long enough for the doctors to have forgotten their earlier measurements. The doctors were also asked to record the length of time it took to complete the measurements.


Part 1

Twenty children (18 boys and two girls) with CP were examined. No parent approached refused consent for their child to take part in the study.

The age range was 3 to 11 years (mean 7 years) and 16 children had severe or profound disability and were non-ambulant with a GMFCS level of 4 or 5. Fifteen children had four-limb involvement (14 had quadriplegia and one had a dyskinesia), four children had a hemiplegia and one had diplegia.

Thirty-two bruises were recorded in 12 children (Fig. 2). Four children had 1 bruise, five had 2, one had 5, one had 6 and one child had 7 bruises.

Figure 2.

Sites of 32 bruises in 12 children.

Of the children with bruises three were ambulant (GMFCS levels 1 or 2). The nine non-ambulant children were level 4 or 5 on GMFCS. One of the eight children without bruises was ambulant.

Not all parents were present when the children were examined. There was an explanation available from child, parent or carer for 13 bruises seen on three children. None of the bruises seen were suggestive of abuse; most bruises were situated on extensor surfaces (shins or back of arms). The children with 5 and 6 recorded bruises had frequent involuntary movements.

All bruises were measured using the template. The smallest bruise was <5 mm by <5 mm, and the largest was 30 mm by 55 mm. Seventeen bruises were 5 mm by 10 mm or smaller.

A variety of other marks were found on 12 children. Many of these were caused by pressure from aids and assistive devices (AFO, boots, standing frames and glasses). Other marks were scars (including in two children who had had chicken pox), and there were a few self-inflicted injuries (i.e. hand biting).

Twelve parents were interviewed by telephone. All comments were positive, especially about the study taking place in school where the children were ‘at home’ and familiar with staff. Six parents expressed the view that it was important to know how disabled children got marks and bruises.

Part 2

Regarding the robustness of the template method, the precision of each technique was estimated by calculating, for each bruise image, the sample standard deviation (SD) of its minimum and maximum diameters as estimated by the various doctors. Averaging over all bruises, it was found that, despite the 5-mm interval in circle sizes in the template, its precision was very similar to that of a standard tape measure, with the mean SDs being 2.9 vs. 3.2 mm (template vs. tape measure) for the minimum and 3.7 vs. 4.1 mm for the maximum bruise diameters. Although, these figures suggest that the template is slightly more precise, the differences are comparable with the statistical uncertainty of the mean SD. It would appear that the principal source of imprecision is the uncertainty in judging where the margins of a bruise lie. Even for simple bruises, this is not straightforward, as a bruise does not have a hard edge.

In our analysis, we have simply used the data as reported. For more complex bruises, further guidance would be helpful. In one particular bruise image, for example, it was unclear to the doctors whether this should be treated as one bruise or three. Unsurprisingly, the variation between observers was then quite large, with SDs of 4.7 vs. 4.8 mm for the minimum and 10.3 vs. 14.3 mm for the maximum diameter. This outlier was included in the aggregate figures quoted above.

The median time required to measure the 25 bruise images was 20 min, whether with template or tape measure, though a few doctors took much longer with the template. Several commented that use of the template became easier with practice.


In our small study examining disabled children for marks or bruises there was no obvious difference in the level of disability between the children with bruises and those without. Of 17 non-ambulant children, nine had bruises and eight did not; the three ambulant children had 1 or 2 bruises each. Two children with involuntary movements had several bruises.

Most of the bruises were small and were situated on extensor surfaces (shins and backs of arms). This is similar to the pattern of accidental bruising seen in the study by Maguire and colleagues (2005) in non-disabled children.

Thirty-six children could have been recruited to this study. Recruitment, however, was slow. Ethical approval was for 1 year and at the end of this time, when 20 children had been examined it was considered that this was a sufficient number to test the method and new measuring technique. There was no reason to think that there was any difference in the two populations, i.e. of the 20 children who were examined and the 16 children who were not examined.

Our study found bruising in non-ambulant disabled children. A recent study of children with neuromotor disabilities (Newman et al. 2010), however, reported that ‘with decreasing mobility there is a clear and significant decrease in the frequency of all injuries especially bruises’.

It is important to consider the use of aids and assistive devices for example AFOs, so that marks can be accurately interpreted.

The use of the template was generally successful and liked by the examiners for its ease of use with the children and comments made included ‘it became quicker and easier to use the template as I got familiar with it’, ‘the template instructions were very precise’ and ‘the circle template was easy to use’.

Most of the imprecision in measurements in the second part of the study arose from the varied nature, shape and colour of the bruises and was very similar irrespective of the technique used. Clearly, if measurements are to be quantitatively representative, a more detailed protocol (for example, including more precise instructions for defining the boundaries of a bruise) would be needed.

Key messages

  • Non-ambulant disabled children can have bruising not associated with abuse.
  • Precise estimation of bruise sizes requires a detailed measurement protocol.
  • A template is at least as precise as a tape measure. It was well liked by examiners and acceptable to subjects and carers.


We would like to thank everyone who helped with this study and in particular:

The pupils and parents who took part;

The management, teaching and healthcare staff in Lancasterian and The Birches Specialist Schools, Manchester;

All community paediatric colleagues who gave their time and energy.