Aim, design, and setting
The aim of this retrospective, cross-sectional, registry study was to compare and explore the use of therapies and interventions prescribed to children with CP, and if these differed by the gender of the child. We also explored if there were differences in treatments received based on if the children were born inside or outside the Scandinavian borders. Foci were on interventions and treatments because these are modifiable in that somebody (provider, caregiver, and/or individual) has made active decisions to prescribe or adhere to specific treatments. Importantly, the aim was not to determine if boys or girls received better treatment (more is not per definition better and in some cases treatments may even be harmful) but to assess, and explain differences in treatments received between genders, if applicable.
The legal caregivers provide oral consent prior to participation in CPUP, and the children provide verbal assent, as applicable. Participation can be discontinued at any time, and the decision to withdraw will not affect the healthcare received. The Regional Ethical Review Board in Lund, Sweden (443-99, revised 2009) approved the study.
Characteristics of participants and description of materials
Data were retrieved from CPUP from the latest physiotherapy and occupational therapy examinations from the years 2016 and 2017. All re/habilitation units in Sweden where children with CP receive care participate in CPUP. Children suspected of having CP are eligible to participate, resulting in a population-based database that includes approximately 95 percent of children with (or suspected) CP born after 2000 (22). The program is highly multidisciplinary and involves professionals from specialties such as orthopedic surgery, pediatric neurology, hand surgery, occupational therapy, and physiotherapy. CPUP, or a modified version of the program, have been implemented in Norway, Denmark, Iceland, Scotland, New South Wales (Australia), and most recently Jordan (22, 23). Participants follow assessment schedules, and based on age and GMFCS level, they complete CPUP assessments once or twice per year. PTs and OTs perform the bulk of the CPUP assessments. These two disciplines are responsible for different parts of the CPUP assessments and the data are entered separately into different forms in the CPUP registry. The characteristics of participants and the distributions of the measures are therefore presented separately for PT and OT treatments and interventions (Tables 1-4).
In total, 2635 participants were included in the analyses on PT related treatments and interventions. The majority, 1528 (58%) were boys and 1107 (42%) were girls. The mean age for boys was 9.5 years (SD = 4.24) and 9.7 years (SD = 4.37) for girls. Furthermore, 3480 participants were included in the analyses on occupational therapy related treatments and interventions. The majority 2014 (58%) were boys and 1466 (42%) were girls. The mean age for boys was 9.7 years (SD = 4.37) and 9.8 years (SD = 4.43) for girls. Separate analyses were made for PT and OT reported treatments, where reports from 2516 individuals overlap and 1082 do not overlap. Of the 1082 non-overlapping reports, 963 were PT and 119 were OT reports.
From the OT form, the following dichotomous treatment modalities (yes/no) were included: use of orthoses in the upper extremities, received additional OT interventions (hand training or activity of daily living training) since the last CPUP assessment or in the last year, and BTX-A in the upper extremities. The local OT recorded the Manual Ability Classification System (MACS) level for each child, describing manual function as one of five levels of how children with CP aged 4-18 years of age uses their hands in daily activities. At level I the child uses the hands easily and successfully, and at level V the child does not handle objects and has severe difficulties in performing simple actions (15). The distributions of the characteristics of the participants and the measures are summarized, by gender, in Table 2.
Statistical analysis
Logistic regression models were used to assess the relationship (presented as odds ratio with confidence intervals) of the outcome variables and gender, adjusted for place of birth, age, GMFCS, MACS (for occupational therapy interventions) and spasticity score (for physiotherapy interventions). Place of birth was dichotomized into Scandinavia (including Sweden, Denmark, Norway and Finland) and outside Scandinavia. The spasticity score was created based on the modified Ashworth scale score for plantar flexors, knee flexors and hip adductors (both left and right side) where ‘1+’ assessment was counted as 1.5. The scores were summarized and divided by maximum total score possible based on the number of sites where the score was available. For example a child with spasticity assessments on all six sites could get a maximum score of 24 (4 points on each site) whereas a child with assessments on three sites could receive a maximum of 12 (children with more than three missing sites assessments were excluded from the analysis). The score was represented as percentage of maximum possible score and the presented odds refer to 1% change in score. The analysis was performed using Stata SE 15.1 (29).