APCP Journal Volume 5 Number 1 (2014) 70-71
A Method to Ease Comparison of Clinical Outcomes in Children with Cerebral Palsy
Penny Butler [*], Richard Major, Pauline Holbrook, Sarah Bew, Lynne Ford
The Movement Centre, Robert Jones and Agnes Hunt Hospital, Oswestry, SY10 7AG
*Corresponding author: email: pennybutler@the-movement-centre.co.uk
________________________________________________________________________________
Background
There is an increasing demand for evidence based
practice in physiotherapy. This demand comes from
clinicians who wish to optimise the management of
patients in their care and from purchasers such as
Clinical Commissioning Groups. When these
questions are directed at paediatric physiotherapists,
and specifically with respect to the treatment of
children with cerebral palsy (CP), there is a variety
of assessment tools that can be used to evaluate and
support a particular treatment approach. One of the
most common is the Gross Motor Function Measure
(GMFM), a standardised and validated measure that
is internationally recognised (Russell et al, 2002).
This evaluative tool provides information about a
child’s functional skills at a specific point in time
and will give some information about abilities
relative to other children with CP. Use of the Gross
Motor Function Classification System (GMFCS)
(Palisano et al, 2007) allows approximate prediction
of functional expectation for an individual child.
The creation of Motor Development Curves
(Rosenbaum et al, 2002) combined the GMFM and
GMFCS and enabled some evaluation of functional
skills relative to the average for a child’s age and
GMFCS Level. The development of reference
percentiles (Hanna et al, 2008a) provided normative
interpretation of GMFM-66 scores for the first time,
further increasing the utility of the GMFM as a
clinical tool. These were provided as graphical data
with tabulated percentiles available on-line (Hanna
et al, 2008b). Clinicians were thus able to make a
more accurate review of a child’s progress against
the average for children in the appropriate GMFCS
Level. The tables provide information at 0, 3, 6 and
9 months for each year from 2 years to 12 years 0
months with GMFM scores enabling identification
of every 5th percentile from 5th to 95th with the
addition of the 3rd and 97th percentile.
As well as providing information about individual
children, these tools offer the potential for
determining how interventions provided by an
individual physiotherapist or by a group of
physiotherapists compare with other
physiotherapists. However, they do not appear to
be used in this way. Research answers the question
70
of ‘is treatment-a better than treatment-b?’ but does
not inform clinicians about their own results
compared to any benchmark. The lack of
comparative data may be because the information is
currently limited to visual inspection of tables and
graphs and to data that is provided in the tables so
that only the value closest to the child’s or group of
children’s age and GMFM score can be used. It is
also difficult to be accurate in plotting a percentile
on a print out of the relevant graph.
This work was therefore to develop a system
extending the use of Reference Percentiles to
numerically calculate the percentile and to be able to
present the mean percentile change over a course of
therapy/intervention so that individual and group
results can be easily and accurately determined.
Method / Results
A two-way linear interpolation1 method of
calculating percentiles was developed, i.e. for age
and for GMFM score. This enabled every possible
combination of age and GMFM score thus
overcoming the limitation of specific age bands and
a limited number of GMFM scores. Software was
further developed so that the user had only to enter
the child’s GMFCS level, age and GMFM score on
an interface screen and the percentile was
automatically calculated. The software also allowed
two data entries so that age and GMFM scores could
be entered at the start and end of a specific time
period. This meant that percentiles could be
determined at yearly review or at start and end of a
specific course of physiotherapy. A further
development was a ‘Report Generator’ that plotted
the information onto the appropriate GMFCS level
percentile graph. This could then be printed and
filed in the child’s clinical notes. The development
of this method also meant that it is possible to enter
the mean GMFM score for a group of children
within the same GMFCS level. This could be done
over a specific time period, e.g. 12 months, using the
mean age of the children at the start and end of the
given period or to determine the average length of a
course of physiotherapy.
P. Butler et al. / APCP Journal Volume 5 Number 1 (2014) 70-71
Discussion / Conclusion
This work has built on the foundation provided by
CanChild and created the potential for percentile
change for a child or the mean percentile change for
a group of children to be easily compared with the
published ‘norm’. Caution must be used when
interpreting percentile comparisons since the
expected within-child variability in percentiles is
substantial. As well as giving a simple and
straightforward method for monitoring the progress
of children, this system could be used to compare
clinical outcomes in children with CP between
physiotherapists, between departments and
possibly even between countries.
References
Hanna, S.E., Bartlett, D.J., Rivard, L.M. and Russell, D.J.
2008b. Reference curves for the Gross Motor Function
Measure: percentiles for clinical description and tracking over
time among children with cerebral palsy. Physical
Therapy.88:596–607.
Hanna, S.E., Bartlett, D.J., Rivard, L.M. and Russell, D.J.
2008b. Tabulated reference percentiles for the GMFM-66 Gross
Motor Function Measure for use with children having cerebral
palsy. [online]. Available at www.canchild.ca [Accessed
June 2013]
Palisano, R., Rosenbaum, R., Bartlett, D. and Livingston,
M. 2007. Gross Motor Function Classification System
Expanded and Revised. [online]. Available at
www.canchild.ca [Accessed June 2013]
Rosenbaum, P.L., Walter, S.D., and Hanna, S.E. 2002.
Prognosis for gross motor function in cerebral palsy: creation of
motor development curves. Journal of the American Medical
Association.288:1357–1363.
Russell, D.J., Rosenbaum, P.L., Avery, L.M. and Lane, M.
2002. Gross Motor Function Measure (GMFM-66 & GMFM88) User’s Manual. London: Mac Keith Press.
______________________________________
1
A method of constructing new data points within the range of a
discrete set of known data points.
71
APCP Journal Volume 5 Number 1 (2014) 70-71
A Method to Ease Comparison of Clinical Outcomes in Children with Cerebral Palsy
Penny Butler [*], Richard Major, Pauline Holbrook, Sarah Bew, Lynne Ford
The Movement Centre, Robert Jones and Agnes Hunt Hospital, Oswestry, SY10 7AG
*Corresponding author: email: pennybutler@the-movement-centre.co.uk
________________________________________________________________________________
Background
There is an increasing demand for evidence based
practice in physiotherapy. This demand comes from
clinicians who wish to optimise the management of
patients in their care and from purchasers such as
Clinical Commissioning Groups. When these
questions are directed at paediatric physiotherapists,
and specifically with respect to the treatment of
children with cerebral palsy (CP), there is a variety
of assessment tools that can be used to evaluate and
support a particular treatment approach. One of the
most common is the Gross Motor Function Measure
(GMFM), a standardised and validated measure that
is internationally recognised (Russell et al, 2002).
This evaluative tool provides information about a
child’s functional skills at a specific point in time
and will give some information about abilities
relative to other children with CP. Use of the Gross
Motor Function Classification System (GMFCS)
(Palisano et al, 2007) allows approximate prediction
of functional expectation for an individual child.
The creation of Motor Development Curves
(Rosenbaum et al, 2002) combined the GMFM and
GMFCS and enabled some evaluation of functional
skills relative to the average for a child’s age and
GMFCS Level. The development of reference
percentiles (Hanna et al, 2008a) provided normative
interpretation of GMFM-66 scores for the first time,
further increasing the utility of the GMFM as a
clinical tool. These were provided as graphical data
with tabulated percentiles available on-line (Hanna
et al, 2008b). Clinicians were thus able to make a
more accurate review of a child’s progress against
the average for children in the appropriate GMFCS
Level. The tables provide information at 0, 3, 6 and
9 months for each year from 2 years to 12 years 0
months with GMFM scores enabling identification
of every 5th percentile from 5th to 95th with the
addition of the 3rd and 97th percentile.
As well as providing information about individual
children, these tools offer the potential for
determining how interventions provided by an
individual physiotherapist or by a group of
physiotherapists compare with other
physiotherapists. However, they do not appear to
be used in this way. Research answers the question
70
of ‘is treatment-a better than treatment-b?’ but does
not inform clinicians about their own results
compared to any benchmark. The lack of
comparative data may be because the information is
currently limited to visual inspection of tables and
graphs and to data that is provided in the tables so
that only the value closest to the child’s or group of
children’s age and GMFM score can be used. It is
also difficult to be accurate in plotting a percentile
on a print out of the relevant graph.
This work was therefore to develop a system
extending the use of Reference Percentiles to
numerically calculate the percentile and to be able to
present the mean percentile change over a course of
therapy/intervention so that individual and group
results can be easily and accurately determined.
Method / Results
A two-way linear interpolation1 method of
calculating percentiles was developed, i.e. for age
and for GMFM score. This enabled every possible
combination of age and GMFM score thus
overcoming the limitation of specific age bands and
a limited number of GMFM scores. Software was
further developed so that the user had only to enter
the child’s GMFCS level, age and GMFM score on
an interface screen and the percentile was
automatically calculated. The software also allowed
two data entries so that age and GMFM scores could
be entered at the start and end of a specific time
period. This meant that percentiles could be
determined at yearly review or at start and end of a
specific course of physiotherapy. A further
development was a ‘Report Generator’ that plotted
the information onto the appropriate GMFCS level
percentile graph. This could then be printed and
filed in the child’s clinical notes. The development
of this method also meant that it is possible to enter
the mean GMFM score for a group of children
within the same GMFCS level. This could be done
over a specific time period, e.g. 12 months, using the
mean age of the children at the start and end of the
given period or to determine the average length of a
course of physiotherapy.
P. Butler et al. / APCP Journal Volume 5 Number 1 (2014) 70-71
Discussion / Conclusion
This work has built on the foundation provided by
CanChild and created the potential for percentile
change for a child or the mean percentile change for
a group of children to be easily compared with the
published ‘norm’. Caution must be used when
interpreting percentile comparisons since the
expected within-child variability in percentiles is
substantial. As well as giving a simple and
straightforward method for monitoring the progress
of children, this system could be used to compare
clinical outcomes in children with CP between
physiotherapists, between departments and
possibly even between countries.
References
Hanna, S.E., Bartlett, D.J., Rivard, L.M. and Russell, D.J.
2008b. Reference curves for the Gross Motor Function
Measure: percentiles for clinical description and tracking over
time among children with cerebral palsy. Physical
Therapy.88:596–607.
Hanna, S.E., Bartlett, D.J., Rivard, L.M. and Russell, D.J.
2008b. Tabulated reference percentiles for the GMFM-66 Gross
Motor Function Measure for use with children having cerebral
palsy. [online]. Available at www.canchild.ca [Accessed
June 2013]
Palisano, R., Rosenbaum, R., Bartlett, D. and Livingston,
M. 2007. Gross Motor Function Classification System
Expanded and Revised. [online]. Available at
www.canchild.ca [Accessed June 2013]
Rosenbaum, P.L., Walter, S.D., and Hanna, S.E. 2002.
Prognosis for gross motor function in cerebral palsy: creation of
motor development curves. Journal of the American Medical
Association.288:1357–1363.
Russell, D.J., Rosenbaum, P.L., Avery, L.M. and Lane, M.
2002. Gross Motor Function Measure (GMFM-66 & GMFM88) User’s Manual. London: Mac Keith Press.
______________________________________
1
A method of constructing new data points within the range of a
discrete set of known data points.
71