DEVELOPMENT OF AN OBSERVATIONAL CHECKLIST FOR ASSESSING HEAD POSITION AND HEAD MOVEMENT IN CHILDREN WITH SEVERE CEREBRAL PALSY

DEVELOPMENT OF AN OBSERVATIONAL CHECKLIST FOR ASSESSING HEAD POSITION AND HEAD MOVEMENT IN CHILDREN WITH SEVERE CEREBRAL PALSY – PRELIMINARY RELIABILITY TESTING

(NORMAL CHILDREN)

C.Honeycombe, A. Ashburn, C.Davey, V.Yule

INTRODUCTION

The position of the head influences essential functions, such as breathing and eating, the regulation of posture, intake of sensory information, ability to participate in activities and communication. Some children with cerebral palsy (C.P.) experience difficulty with head control. Special seating is a key approach for the management of impaired head control, but difficulties in this aspect can remain. Despite the significance of head position there is a lack of research in this area. A detailed profile of the way in which a child moves and positions their head could assist in the selection and evaluation of seating intervention. Existing seating evaluation scales (Fife 1991, Myhr 1991, Reid 1995) do not contain sufficient information to enable a detailed profile of head position and movement to be developed. Consequently a checklist for the assessment of head position and head movement in children with C.P was developed. Initial stages of development involved collecting information about head position from four sources, the literature, interviews with paediatric therapists, researchers experienced in movement analysis and all day observations of ten children with impaired head control. This information was subjected to content analysis, independently by two researchers, a process which resulted in an 18-item video based categorical checklist. To enable development of a small database of head position/head movement in children without disabilities, initial checklist testing was on video data from 20 normal children. Findings will be used to assist in the interpretation of subsequent data collected from children with C.P.

AIMS OF RELIABILITY TESTING

· To explore intra and inter rater reliability of the checklist using videos of normal children

· To record head position and head movement in normal children using the checklist

METHOD

Filming - Twenty children without disabilities were filmed in a controlled environment undertaking three different activities, playing a computer game, listening to a story and having a conversation. Each activity was filmed for 3 minutes. The camera was positioned to the left side of the child perpendicular to the flexion and extension plane for head movements. The same chair and table was used for all children but smaller children also used a foot block. A laptop computer was positioned directly in front of the child at a comfortable distance. The researcher sat opposite the child when reading the story and during the conversation. The computer game involved scanning the screen and selecting items with the mouse. The questioning was designed to encourage more than yes/no answers to enable video recording during longer periods of the child talking.

Checklist description - The checklist comprises of six broad categories, dominant head position, contact with head support, trunk involvement, head stillness, lack of head control and head activity. It was designed for completion after watching a 3-5 minute video of a child filmed in the sagittal pane. A photographic picture reference system (15 positions seen in children with C.P.) is included for the rating of head position/s. Items requiring estimates of the duration of behaviours are rated from a choice of 90%+, 50-89%, or <50%. Sections on trunk involvement and lack of head control require selection from written categorical options.

Procedure for testing inter and intra rater reliability of the checklist - Written and verbal instructions on how to complete the checklist were given to three senior therapists undertaking the reliability testing. For inter-rater reliability testing these therapists watched each video clip and independently rated head position and head movement using the checklist. Each rater completed three checklists per child, one for each activity period i.e. playing on a computer, listening to a story and during the conversation. For intra –rater reliability one therapist rated the videos a second time, one-month after the first rating. Data were analysed using the kappa statistic and by examining the frequency of agreements and disagreements. Agreement and disagreement figures will be presented as kappa scores were difficult to interpret. Checklist ratings tended to cluster around a single point and Kappa is influenced by the prevalence of observations in each category (Altman 1991). In such circumstances it is recommended that greater emphasis be placed on raw data (Carr et al 1999). High agreement is indicated where agreement reaches 70% or more (Carr et al 1999). In this analysis high agreement for a checklist item was indicated where all three therapists selected the same coding for 70% of the children filmed (n=14).

RESULTS

Twenty children were filmed. They had a mean age of 9.2 years with a range of 5-15 years.

Table 1. Picture reference items, frequencies where all ratings agreed for an individual child (*high agreement)

CHECKLIST ITEM	Computer		Listen		Conversation
CHECKLIST ITEM	Inter rater	Intra rater	Inter rater	Intra rater	Inter rater	Intra rater
Picture dominant head position	9	17*	7	17*	11	15*
Extent of head movement (maximum)	2	14*	8	16*	6	16*
Predominant range of head movement	7	16*	3	14*	7	11

Ratings using the picture reference system showed high intra rater agreement, with the exception of the predominant range of head movement during conversation. However high inter rater agreement was not achieved. Examination of inter rater disagreements showed most were between adjacent categories e.g. upright head/upright head +chin jut.

Table 2. Estimation of duration of behaviour items, frequencies where all ratings agreed for a individual child (*high agreement)

CHECKLIST ITEM	Computer		Listen		Conversation
CHECKLIST ITEM	Inter rater	Intra rater	Inter rater	Intra rater	Inter rater	Intra rater
Proportion of the assessment time that the dominant head position was held	11	17*	10	15*	7	17*
Maximum time head still	16*	19*	17*	20*	5	10
Frequency of head movement	10	17*	14*	15*	2	15*

Good intra rater reliability was achieved for all items, based on estimations of the duration of behaviours, except for the maximum time that the head was held still during talking. Good inter rater reliability was achieved for the maximum time that the head was still during the computer task and listening and for the frequency of head movement during listening.

Some checklist items were not applicable for children without disabilities, contact with head support, lack of head control, and smooth or jerky movement. Others only underwent limited testing as they were demonstrated in a very small number of children, dominant movement pattern, proportion of time that the dominant pattern was demonstrated. Reliability results for these are not presented in this paper.

DISCUSSION - The development of the checklist (content, validity and reliability) has identified important domains of head position and head/movement in children with head control impairments. Intra rater reliability of the items was high under most conditions but inter rater reliability was only high for two items. The checklist was designed for use with children with head control difficulties therefore the results of inter rater reliability with normal children cannot be taken as a representation of how the tool will perform with children with C.P. Documentation of head position/movement based on observation is a challenging task and other researchers (Fife 1991 and Carr 1999) have also experienced difficulties with demonstrating reliability. Much of the inter rater disagreement was between adjacent categories within an item. The reliability of the picture reference items may improve when used with children with C.P. as the pictures may more closely match the positions of the children. Some disagreement appears to have resulted in the use of different thresholds by raters for coding whether or not movement has occurred, despite attempts to minimise this through written descriptions. The higher intra rater reliability indicate that the incorporation into the checklist training of illustrative videos of movement or non- movement examples may improve reliability for items dependent on estimations of the extent or duration of movement.

FUTURE WORK - The checklist will be tested with a group of 10 children with C.P. and difficulty maintaining upright an upright head. External validation of two checklist items will be determined using 2D video motion analysis. The findings for this will also be used to create a database of head position/ movement in children with C.P. and normal children.

ACKNOWLEDGEMENT

Research was provided by the NHS Executive South East Research Training Fellowship.

REFERENCES

1. Fife S.E., Roxborough L.A., Armstrong R.W., Harris S., Gregson J.l., & Field D. Development of a Clinical Measure of Postural Control for Assessment of Adaptive Seating in Children with Neuromotor Disabilities. Physical Therapy 1991;71:981-993.

2. Myhr U. & Lennart von Wendt. Improvement of Functional Sitting Position for Children With Cerebral Palsy. Developmental Medicine and Child Neurology 1991;33:246-256.

3. Reid D.T. Development and preliminary validation of an instrument to assess quality of sitting of children with neuromuscular dysfunction. Physical and Occupational Therapy in Paediatrics 1995;15:53-81.

4. Altman D.G. Practical statistics for medical research. London, Chapman and Hall, 1991.

5. Carr E.K., Keney F D., & Wilson-Barnett J.N.D.J. Inter-rater reliability of postural observation after stroke. Clinical Rehabilitation 1999;13:229-242.

Notes: