Infants have malleable and fast-growing cranial bones, and are therefore at risk of developing skull deformation if their head often remains in the same position. When a child turns its head toward one side most of the time, this is defined as positional preference . Skull deformation due to such prolonged external forces (non-synostotic) must be distinguished from skull malformation due to premature fusion of the cranial sutures (synostotic) . Deformational brachycephaly refers to a symmetric occipital flattening of the skull that is sometimes accompanied by temporal bossing or an occipital lift . The term deformational plagiocephaly is used to describe a unilateral occipital flattening of the skull. More severe cases often present with ear misalignment and facial asymmetry [2, 4].
Skull deformation is generally considered a purely cosmetic disorder. Yet parents worry that the deformation might be permanent and might influence the child’s attractiveness with the risk of, for example, being teased . Some studies suggest long-term developmental delays due to skull deformity, but no causal relationships have been found [5–7].
The prevalence of skull deformation can be up to 21.5% in infants younger than 6 months, but decreases within the first years of life [8–10]. A low parental level of education, ethnicity, male gender, primiparity, prematurity, birth factors, delayed (motor) development, low activity level and several positioning and dietary factors have been reported as risk factors, while placing a child in the prone position when awake appears to be a protective factor [1, 11–17].
Prevention or treatment of positional preference and skull deformation include parental counselling, counter-positioning and physical therapy [10, 18]. Children with persisting severe skull deformation at the age of 5 to 6 months are commonly treated using orthotic devices (redression helmets or headbands) [4, 19]. In The Netherlands, a redression helmet costs about €1,200 and is reimbursed by health insurance companies as well as the accompanying visits to the (paediatric) physician. However, until now, no randomised controlled trials (RCTs) have been performed to study the effectiveness of this therapy [20, 21]. The few non-randomised studies tend to show positive results, but have several limitations. To start with it is unknown whether the reported differences in effectiveness are clinically relevant. Furthermore, follow-up in these studies was short-term (either directly after treatment or just a few months afterwards), there was a lack of blinding or information about blinding and often no validated outcome measures were used. Finally, data about complications were not collected in a structural way in these studies [2, 4, 20, 22, 23]. Although the known complications of helmet therapy are mild and do not seem to occur often, the treatment burdens both parents and their young children . Next to the lack of scientific evidence, experience shows differences in beliefs and referral policies of health care professionals regarding helmet therapy. Some advocate the use of helmets to treat skull deformation, while others are reluctant to prescribe this intensive treatment for a cosmetic condition without knowing its effectiveness [21, 24]. This makes parents very uncertain when they have to decide whether to start helmet therapy or not.
Both helmet therapy and no helmet therapy (allowing natural recovery) are standard approaches in The Netherlands. To compare the effectiveness of these two approaches a pragmatic RCT study design is required . Pragmatic trials are designed to find out the effectiveness of a treatment in routine, everyday practice and thereby have a high external validity [26, 27]. A high external validity can be achieved by recruiting a broad study population that is representative of the target population, studying interventions that approach a real world delivery of care, applying blinding to neither participants nor specialists and selecting a wide range of outcome measures [28, 29].
Since the condition of interest changes over time and the decision-making is time-dependent, the RCT needs to be nested in a cohort study [30–32]. The decision to start helmet therapy is usually taken at 5 to 6 months of age. Recruitment at that stage is complicated as the children tend to be scattered among various institutes if their parents prefer helmet therapy or are outside the health care system if their parents choose not to start helmet therapy. As the cohort study recruits children at risk of disease progression before helmet therapy can be prescribed, we tackle this problem and we are also able to predict the number of children that ultimately will be eligible for helmet therapy and identify prognostic factors.
Additionally, nesting the RCT in a prospective cohort study makes it possible to present information on the representativeness of the RCT population, by comparing this population with non-participants [33, 34]. Furthermore, outcomes of the randomised trial can be compared with the parallel non-randomised trial that employs the same types of intervention.
The main goal of the Helmet Therapy Assessment in Deformed Skulls (HEADS) study is to investigate the effects and costs of six months of helmet therapy compared to no helmet therapy in children with moderate to severe skull deformation. This article describes how this study is designed and reports the recruitment scheme so far. We provide a description of the statistical analysis plan to be used after data collection is completed and conclude with general recommendations on study design.