This review examines the research on the Picture Exchange Communication System (PECS), which has become a popular communication strategy for children with autism and other communication disorders. A growing body of research has shown that the system is a promising mode of communication. There is, however, a paucity of research that examines the conditions under which the PECS is used, specifically whether children use the PECS to make spontaneous requests. A lack of agreement currently exists over the definition of the term ‘spontaneity’ and so researchers of the PECS who do report instances of spontaneity may be basing the judgment on different patterns of behaviour. Skinner’s (1957) analysis of verbal behaviour and the continuum model of spontaneity (Carter, 2002,2003a; Carter & Hotchkis, 2002; Chiang & Carter, 2008) can be used to understand the development of self-initiated requesting behaviour. Both frameworks state that requests can only be considered as fully spontaneous if they occur without prompts from another person and when the desired item is not in sight. There is a lack of research that examines whether children are able to use the PECS to make requests under these conditions. Furthermore, this literature review shows that some children may be unable to use the PECS to request items not in sight because of the teaching conditions used and/or because the reinforcement practices of the community may be inefficient, and, therefore, ways of promoting spontaneity are considered.

No description available.

Fundamental movement skills (FMS) are a vital part of a child’s development, which allow them to advance on to more complex movements, resulting in them functioning successfully in their daily lives and when participating in sports and physical activity. Despite the importance of FMS, previous studies have routinely found children to have low competency levels. Further research on competency levels in this area is required, specifically among UK children, as data on FMS proficiency and FMS interventions is dominated by research from Australia and North America. As such, the aims of this thesis were to a) document the level of FMS competency of preschool children from a highly deprived area of Northwest England, b) determine the effectiveness of a six-week Active Play intervention on FMS competency among preschool children from a highly deprived area of Northwest England, c) examine the relationship between FMS competency, physical activity and weight status over a five-year period between preschool and late primary among children from a highly deprived area of Northwest England and finally, d) gain the thoughts and opinions of experts and practitioners in order to help inform the development of an appropriate intervention to increase the physical literacy of preschool children.

Disorders of sex development (DSD) are a diverse group of conditions in which there is variation from the typical chromosomal, gonadal, or anatomical developmental pathway. While much has been learnt about the genetic aetiology of many of these disorders, a significant proportion of cases remain without a definitive diagnosis. This thesis consists of a series of studies designed to look at different aspects of DSD in order to identify causes and develop better ways to assess and research these conditions in the future. Chapter 1 is an extensive literature review of normal sex development, models to describe the sex phenotype, steroidogenesis, steroid hormone structure and physiological role, classification of DSD, clinical uses of progestogens and determinants of foetal growth. An understanding of these diverse subjects is essential to consider the topics investigated. Chapter 2 presents the rationale for, and specific aims of, this thesis. Chapter 3 describes a study using the I-DSD registry, the largest international register of cases of DSD, to identify associated conditions co-occurring in DSD. 649 cases with documented consent were identified and analysed from the registry, with further information obtained from the reporting clinician where necessary. Associated conditions were reported in 168 cases (26%), overall, and when considered according to karyotype were reported in 112 cases of 46,XY DSD (24%), 27 cases of 46,XX DSD (22%), 19 cases of 45,X/46,XY (45%), and 6 cases of 45,X (75%). In 46,XY DSD, which represents the largest group of cases in the Registry, small for gestational age (SGA), cardiac and CNS anomalies were the most commonly reported associated conditions. This study strengthens the recognised association between SGA and non-specific 46,XY DSD. Additionally, the data indicate a possible association between genetically confirmed AIS and skeletal and renal anomalies. These results provide new research targets for cases in which the aetiology of DSD remains unclear. They also highlight the need for multi-disciplinary teams for management of these patients. Chapter 4 documents a clinical study investigating the association between hypospadias, one of the mildest conditions on the spectrum of DSD, and the measurement of anogenital distance (AGD). 88 boys had AGD measured under general anaesthetic during hypospadias surgery, alongside assessment of severity of hypospadias. Medical notes were reviewed for further information including birth weight, gestation and the presence of additional genital anomalies, as described by the external masculinisation score (EMS). Median AGD was found to be shorter in boys with severe hypospadias (63mm), than those with mild hypospadias (75mm) (p < 0.001). Additionally these boys were more likely to have lower birth weight SDS (-0.61) than boys with mild hypospadias (-0.42) (p= 0.013). This study is the first to show a link between severity of hypospadias, additional genital anomalies, and degree of AGD shortening. This supports the hypothesis that severe forms of hypospadias may be linked to inadequate androgen exposure in utero. Results also show that boys with more severe hypospadias have a lower birthweight, reinforcing the link between 46,XY DSD and SGA. Chapter 5 describes the use of a rat model to investigate the developmental effects of exposure to the progestogen medroxyprogesterone acetate (MPA) during the male programming window. It has been previously suggested that antenatal exposure to progestogens may be associated with DSD. In this study Sprague Dawley dams were injected with 75mg/kg or 150mg/kg of subcutaneous MPA on gestational days 14.5 to 18.5. Results showed that MPA exposure was associated with a shorter than normal AGD in male rats, and a longer AGD in female rats. Offspring of both sexes had reduced birth weight when exposed to MPA (control weight 5.99g, MPA 75mg/kg 4.58g, MPA 150mg/kg 4.72g). There was no evidence of an effect on internal reproductive structures, including testis weight. Chapter 6 describes studies using small vessel myography to investigate vascular function in the pregnant dams exposed to MPA in the previous studies. Low birth weight can be the result of altered vascular remodelling during pregnancy, leading to impaired placental function. It has previously been suggested that impaired placental function may be responsible for the combination of intra-uterine growth restriction and DSD. Uterine artery segments from animals exposed to MPA 150mg/kg demonstrated greater vessel wall thickness, and a trend towards an increase in internal and external diameter, with increased distensibility at higher pressures when compared to control segments. Wire myography showed that vasoconstriction in response to noradrenaline and NG-nitro-L-arginine methyl ester (L-NAME) was reduced following MPA exposure. These studies do not explain the causes of reduced growth in MPA exposed offspring. The responses seen are the opposite of those seen in animals with pre-eclampsia and hypertension, and may demonstrate the protective effect of progestogens in pregnancies complicated by these conditions. Chapter 7 draws together the findings of all the studies in this thesis, to reach overall conclusions. The common theme of an association between DSD and impaired foetal growth from all three branches of this work in discussed. The potential for further investigations in pursuing this work to strengthen conclusions and inform future practice is considered.

The increasing numbers of children with autistic spectrum disorders (ASD) being educated within mainstream schools and the deployment of Teaching Assistants (TAs) to work with them has resulted in significant challenges in relation to the support and training TAs require when working with pupils who have ASD. Whilst there is a significant body of literature relating to the development of specific interventions for children with ASD there has been little that looks at the relationships formed between professionals, particularly Teachers and TAs, and these pupils. In order to extend the limited evidence base on the relationships that TAs form with pupils who have ASD this research explored the experiences of a group of six TAs working directly with pupils who had ASD and were in mainstream primary schools, and the relationships that they formed with them. The data collected from the TA interviews was analysed using Interpretative Phenomenological Analysis (IPA) and subordinate and superordinate themes identified. The findings were then discussed in the context of Interdependence Theory, particularly the investment model put forward by Rusbult and Buunk (1993). The study found that TAs described their relationships with the children through four main superordinate themes, that is, attachment to the relationship; the difficulties presented by the child; the position they took as TA in the relationship and the personal and professional satisfaction they got from the relationship. The data from the transcripts showed that TAs invested significant amounts of time, effort and energy into developing their relationships and this resulted in rewards and costs for the TAs. These contributed to how satisfied the TAs were with the relationships and how committed they were to them. The implications for TAs, Schools and Educational Psychologists were discussed, as were the implications for future research.

Cardiovascular disease is one of the largest killers in the UK representing 30% of all global deaths. The underlying processes of the disease are thought to begin in childhood. Whilst traditional risk factors of CV disease (e.g. hypertension, hyperlipidemia, obesity, smoking, stress and sedentary lifestyles) are becoming increasingly prevalent within the younger generation there remains the need for the establishment of earlier or "pre-clinical" markers of future CV disease risk or current atherosclerotic load such as left ventricular (LV) mass, diastolic function and carotid intima-media thickness( cIMT). Further, assessing the association of these markers to other risk factors and then determining the impact of physical activity (PA) interventions is warranted. Initially we assessed the impact of body composition, PA and cardiorespiratory (CR) fitness upon left ventricular LV mass, carotid-intima media thickness (cIMT) and LV diastolic function in 218 9-11 year old primary schoolchildren. Pubertal status was assessed through a maturity offset calculation. LV mass, cIMT, and LV diastolic function were assessed via ultrasound. Body mass index was assessed via anthropometry whilst fat mass [FM] and lean mass [LM] were determined via dual X-ray absorptiometry. Average 3-day PA was recorded via a uni-axial accelerometer and CR (VO₂peak) was determined from a graded treadmill test. Relationships were analysed using bivariate correlations and forced entry multiple regression. All children were classified as being below their peak height velocity. Together LM, FM, sex and moderate to vigorous (MV)PA accounted for 59% of the variance in LV mass with LM being the most important predictor (P<0.005). Sex, LM, FM and VO₂peak explained only 19% variance in cIMT and just 9% of the variance in LV diastolic function was accounted for by LM, FM blood pressure and sex. Data for MVPA had no significant relationship to any cardiovascular (CV) variables although was negatively correlated with FM. The strong association between LV mass and body composition likely represents normal growth. The limited shared variance between predictor variables and cIMT and LV diastolic function suggests that those pre-peak height velocity children in the current cohort who were overweight, inactive and unfit were not yet at an increased CV disease risk. Thus there is a window of opportunity for intervention programmes to be implemented that reduce CV disease risk before adolescence and adulthood. Following this, an exploratory trial was conducted to introduce the use of PA interventions in pre-pubertal children. Sixty-one 10-11 year old Liverpool primary school children volunteered and were randomly assigned by school to a STEX programme (2 x 60 min sessions per week at a heart rate of ~145 beats min⁻¹), a PASS programme (weekly physical activity tasks and pedometer challenges) and a control (CON; no intervention). Pre-clinical CV measures and body composition were measured before and after the 9-week intervention period. The primary outcome variable was cIMT, with LV mass, LV diastolic function, and body composition defined as secondary outcomes. Delta (Δ) scores were analysed by ANCOVA, with baseline scores as the covariate. For the primary outcome, the probability that the population effect of the intervention is at least as great as the pre-specified minimum clinically importance difference (MCID) was estimated, to evaluate clinical relevance. All participants met 75% compliance criteria for STEX and PASS. The effect of the STEX intervention (compared with CON) was a mean benefit of -0.018mm for average maximum cIMT (90% CI, -0.039 to 0.002mm), and -0.016mm for average mean cIMT (90 % CI, -0.040 to 0.008mm). The probability (% chances) that the true population effect of the STEX intervention would be clinically beneficial was 79% for average maximum and 71% for average mean cIMT. The PASS intervention did not result in clinically important effects, and no other substantial changes were observed for the secondary outcome variables. The relatively high probability of clinically beneficial effects of the STEX intervention suggests that a larger, "definitive" randomised trial with longer follow-up is warranted to define the effectiveness of the intervention more precisely. As a consequence the longer PA intervention study observed 152 children aged 9 to 10 years over 12 months. All of the echocardiographic, body composition, CR fitness and PA variables mentioned were assessed as previously discussed. Children were randomly assigned by school to an intervention group. Control (no intervention), PASS (as before except delivered during school hours to enhance compliance), high intensity physical activity (the same as STEX but renamed due to the addition of another structured exercise group) or fundamental movement skill ([FMS] 2x 60 min sessions per week of skill based activities). These interventions took place over a year period with participants being assessed at baseline, approximately mid-way through and post-test (52 weeks). Initial factorial ANOVA analysis comparing all 3 intervention groups and a control group before, during and after the 12 month intervention period, found limited statistically significant evidence for a positive impact of PA interventions compared to controls in pre-pubertal children. However, after adjusting for confounding variables in an ANCOVA analysis some sporadic benefits of PA interventions on CV variables were uncovered. An increase in LV mass over 12 months, after adjustment, was lower in the HIPA group compared to CON group (11.5 g; 90% CI, 2.0 to 21.0 g). This change was also lower in the FMS group compared to CON group (13.8 g; 90% CI, 4.6 to 23.1 g). The ANCOVA adjusted change scores for both mean and max cIMT were less in the intervention groups compared to CON group but only in the PASS group were these differences significant (P<0.05). PASS increased its mean cIMT (-0.014mm less than control (90% CI, -0.002 to -0.030)). Somewhat surprisingly the intervention programmes had no positive effect on CR fitness (indeed this decreased), PA measures and/or body composition over and above changes observed due to growth. This thesis has provided a unique insight into the 'pre-clinical' CV disease risk factors in pre-pubertal children and the impact of differing PA activity interventions with this group. Interestingly the research has shown that within this population overweight/obese, inactive low CR fit individuals are generally not at a higher CV disease risk than their aged matched 'healthier' counterparts. When PA interventions are introduced in the short term positive changes in cIMT were seen, however, this is not reciprocated in longer PA interventions possibly due to a larger maturation effect over 12 months. Interestingly year long interventions provide some attenuation of growth-related changes in CV disease risk factors but these changes are generally small and sporadic. It is suggested that further research over a longer period of time with more 'at risk' populations is needed. The PA interventions adopted achieved high attendance and compliance records and thus may be transferable out of the research process. It is interesting to also speculate that future research may not need to administer high impact activity, as previously thought, as some positive data was obtained in more general lifestyle interventions involving more knowledge transfer.

Physical activity (PA) and reducing the time spent sedentary can favourably impact the health and wellbeing of preschool aged children. Despite the importance of regular P A, objective monitoring studies suggest that young children do not accumulate enough P A to benefit their health. Furthermore, relatively few interventions to promote PA in children under 5 have been rigorously evaluated and intervention studies to date have generated mixed results. The aims of this thesis were to a) assess within-day variability of objectively measured moderate-to- vigorous physical activity (MVP A) during weekdays and weekend days among preschool children; b) investigate the effect of a curricular Active Play intervention on children's sedentary and P A behaviours; and, c) investigate the effect of a family focused Active Play intervention on children's weekday and weekend day sedentary time and total P A. Study 1 Background: Identifying periods of the day which are susceptible to varying levels of P A may help identify key times to intervene and potentially change preschool children's PA behaviours. This study assessed variability of objectively measured MVP A during weekdays and weekend days among preschool children. Method: One hundred and eighty-eight children (aged 3-5 years; 53.2% boys) from a northwest English city wore uni-axial accelerometers for 7 consecutive days. Results: Higher levels of MVP A were recorded in boys, particularly those who attended preschool for a half day. Children who attended preschool for a full day engaged in 11.1 minutes less MVPA than children who attended for a half day. After-school hours were characterised by a decrease in activity for all groups. Patterns of activity during the weekend were smoother with less variability. Conclusion: This study identified discrete segments of the week, specifically afterschool and during the weekend, when preschoolers engage in low levels of PA. Higher levels of MVPA among children who attended preschool for less time each day suggests that the structured preschool environment is related to decreased activity. Consequently, there is a need for interventions in young children to focus on school and home environments. - Study 2 Background: Early childhood is a critical time for the promotion of healthy lifestyle behaviours such as physical activity. Cross-sectional studies suggest that preschool children are not sufficiently active to benefit their health. Few studies have investigated the effect of interventions on physical activity levels and time spent being sedentary in this population. The aim of this study was to investigate the effect of a school-based active play intervention on preschool children's physical activity and sedentary time. Method: Two hundred and forty children were recruited from 12 randomly selected preschools in the North West of England. Six schools were randomly assigned to either an intervention (n=6) or a current practice comparison group (n=6). One teacher in each intervention school received training in the delivery of a 6-week active play programme including active play resources. The training programme was delivered by experienced active play professionals. Comparison schools received the active play resources (but not the training) and were asked to deliver their usual active play provision. Children wore a uni-axial accelerometer for 7 days at baseline, post-test and at 6 months follow up. Whole week sedentary time and physical activity adjusted for pupil- and school- level covariates were analysed using multilevel analyses. Results: No significant intervention effects were observed for sedentary time or physical activity. Sex and hours spent at school were significant predictors of physical activity in this age group, with boys engaging in less sedentary time and more light intensity activity than girls. Children who spent fewer hours at school were significantly more active than those who spent more time at school. BMI, maternal education, child's ethnicity and accelerometer wear time were not associated with activity levels. Conclusion: Physical activity during intervention classes was high. However the six teacher training intervention sessions did not increase physical activity or decrease sedentary time over a week. Specific covariates were identified as having a significant effect on the outcome measures. Moreover, children who spend more time at pre-school were less active suggesting that this setting was not as conducive to physical activity engagement as other environments. Study 3 Background: Early childhood provides a window of opportunity for the promotion of physical activity. Given the limited effectiveness of interventions to date, new approaches are needed. Socio-ecological models suggest that involving parents as intervention targets may be effective in fostering healthier lifestyles in children. This study describes the effectiveness of a family-focused' Active Play' intervention in decreasing sedentary time and increasing total physical activity in preschool children. Method: Seventy-seven families were recruited from 8 randomly selected SureStart children's centres in the North West of England. Centres were randomly assigned to either an intervention (n=4) or a comparison group (n=4). Parents and children in the - intervention group received a IOeweek active play programme delivered by trained active play professionals; this included an activity and educational component. Families in the comparison group were asked to maintain their usual routine. Each participating parent and child wore a uni-axial accelerometer for 7 days at baseline and post-test. Week and weekend day sedentary time and total physical activity adjusted for child- and home- level covariates were analysed using multilevel analyses. Results: Significant intervention effects were observed for sedentary time and physical activity for both week and weekend days. Children in the intervention group engaged in 1.5% and 4.3% less sedentary time during week and weekend days, respectively and 4.5% and 13.1 % more physical activity during week and weekend days, respectively than children in the comparison group. Parent's participation in sport and their physical activity levels, child's sex, availability of media in the home and attendance at organised activities were significant predictors of sedentary time and physical activity in this age group. Conclusion: A 10-week family focused active play intervention produced positive changes in sedentary time and total physical activity levels in preschool children. Specific covariates were identified as having a significant effect on the outcome measures. Moreover, children whose parents were active engaged in less sedentary time and more physical activity suggesting that parent's activity habits are mediators of physical activity engagement in this age group.

Objective The aim of this study was to examine the effects of selected routine nursing interventions - endotracheal suctioning and manual ventilation (ETSMV), log-rolling, eye care, mouth care and washing - on the intracranial pressure (ICP) in children with traumatic brain injury. Design Prospective observational study over three years. Setting Single tertiary paediatric intensive care unit in the North West of England. Patients Twenty five children with moderate to severe closed traumatic brain injury and intraparenchymal intracranial pressure monitoring in intensive care (2 -17 years of age). Interventions Routine nursing care interventions. Measurements and main results ICP measured one minute before the procedure, at the maximal value during the procedure and five minutes after the procedure was recorded for the purpose of this study. Time to recovery was also recorded, in minutes. A total of 25 measurements (the first one in each child) in the first 36 hours of the child's PICU admission were analysed. Both ETSMV and log-rolling were associated with clinically and statistically significant changes in ICP from baseline to maximal ICP (p=0.005) and maximal to 5-minute post ICP (p=0.001) for ETSMV and (p < 0.001) baseline to maximal ICP and (p=0.002) for maximal to post-procedure ICP for log-rolling. During ETSMV and logrolling 70% of children exceeded the 20mmHg clinical treatment threshold during the interventions. During both ETSMV and log-rolling children with higher baseline ICPs ( > 15mmHg) showed higher maximal ICPs (but not ICP rise), suggesting a linear relationship between baseline and maximal ICP, although this was more pronounced during turning. One third of the children had not returned to their baseline ICP by 5 minutes after ETSMV, compared with 60% children after log-rolling. Neither eye care nor mouth care showed any clinically significant effects on ICP in these children, suggesting these procedures are not noxious and are tolerated very well. However, there was only a small number of washing episodes reported in this study therefore the observations are not conclusive. Conclusions Endotracheal suctioning and log-rolling in moderate to severe traumatic brain injured children can cause significant intracranial instability and should only be performed as required and with careful planning and execution. Eye and mouth care and washing appear to be well tolerated interventions and could be performed when necessary.

The aim of this thesis was to investigate the impacts of physical activity (PA), cardiorespiratory fitness (CRF), body composition and lifestyle education on cardiometabolic (CM) risk in 10-11 year old children. This broad aim was approached using three studies. Studies 1 and 2 were cross sectional observational studies, and study 3 was a clustered randomised control trial, with intervention effects assessed at post intervention and again at 8 to 10 weeks after the intervention. Initially, in the first cross sectional study (Chapter 4) the relationships between non- invasive (LV Mass, E/A, E'/A', E/E', trunk fat mass, whole body fat mass) and invasive CM risk markers (CRP, HOMA-IR, adiponectin, TC: HDL-C), and between all risk markers and CRF (V02 peak), time spent sedentary, moderate to vigorous intensity PA (MVPA) and vigorous PA (VPA) were investigated in 10-11 year old children (n=62). The key findings were significant but generally weak relationships present between some of the non- invasive and invasive markers of CM risk and risk markers also had significant correlations with measures of CRF and PA. CRP was significantly positively correlated with whole body fat in boys (rho = 0.486, p < 0.05) and girls (rho = 0.485 , p < O.Ol) and with trunk fat mass in boys (rho = 0.384, p < 0.05) and girls (rho =0.489, p < O.Ol). Adiponectin was negatively correlated with whole body fat (rho = -0.446, p < 0.05, and R=-0.697, p < O.Ol) and trunk fat mass (rho = -0.614, p < O.Ol; rho = -0.475, p < O.Ol) in boys and girls respectively, and in girls adiponectin also correlated positively with E'/A' (r=0.356, p < 0.05). In boys only, TC:HDL-C was positively correlated with whole body fat (rho =0.407, p < 0.01) and trunk fat mass (rho =0.391, p < 0.05). ; V02Peakwas negatively correlated with CRPin boys (Rho = -0.492, p < 0.05) and HOMA-IR in girls (Rho =-0.522, p < 0.01). V02Peak was also negatively correlated with whole body fat (rho =-0.515, p < 0.01; r=-0.697, p < O.Ol) and trunk fat mass (rho=-0.494, p < 0.05; rho =-0.706, p < 0.01) in boys and girls respectively. Both MVPA and VPA correlated negatively with TC: HDL-C in girls (rho= -0.396, p < 0.05; rho =-0.428, p < 0.05) and MVPA correlated with whole body fat (rho= -0.602, p < O.Ol) and trunk fat mass (rho=-0.65, p < 0.01) in boys. VPA also correlated with whole body fat in girls (rho= - 0.544, p < O.Ol) and with trunk fat mass in both boys (rho= -0.428, p < 0.05) and girls (rho= - 0.468, p < 0.01). Time spent sedentary had a positive correlation with whole body fat in boys (rho = 0.429, p < 0.05). This study demonstrated that risk factors clustered in individuals and that relationships were present between invasive and non-invasive markers of cardiometabolic risk, and provided preliminary evidence to investigate this phenomenon further. The correlations described in this study suggest a clustered risk score which includes both invasive and non-invasive measures may add value to predicting overall risk. The second cross sectional study (Chapter 5) investigated clustered CM risk, by combining invasive markers with non-invasive 'pre-clinical' markers of CM risk into a clustered risk score, in a different cohort of 10 - 11 year old children. Clustered risk scores were negatively correlated with CRFand PA. V02 peakshowed a moderate negative correlation with CRSA (r=-0.57, p < 0.01) and CRS B (r= -0.60, p < O.Ol) VPA showed a moderate negative correlation with CRSA (r= -0.51, p= 0.01) and CRSB (r= -0.50, p=O.Ol). MVPA showed a moderate negative correlation with CRSA (r= -0.44, p= 0.03) and CRSB (r= -0.41, p=0.04). Sedentary time showed a moderate positive correlatcorrelation with CRSA (r= 0.414, p= 0.049). The evidence provided by these two observational studies, Study 1 (Chapter 4) and Study 2 (Chapter 5), along with other literature, as discussed throughout this thesis, gave rationale for an intervention with the aim to reduce negative lifestyle behaviours, of low levels of PA, high levels of sedentary behaviour and poor nutritional balance, increase CRF and maintain a healthy body weight.

Background: Sedation withdrawal is one of the terms used to describe the behavioural response to stopping or reducing sedative drugs in physically dependent patients. Withdrawal behaviours differ according to the drug involved and may be unpleasant and interfere with recovery. Recognition of sedation withdrawal is challenging due to differences in patient presentation and may be further complicated by the patient’s condition and concomitant drug therapy. Overall Aim of the full thesis: To improve the accuracy of sedation withdrawal assessment in critically ill children. Objectives and Methods: A mixed methods interactive approach comprising six studies. Study 1 evaluates the psychometric properties of the Sedation Withdrawal Score, Studies 2 and 3 examine the complexities/challenges of withdrawal assessment by critiquing existing tool validation studies, A further three studies examine the nurse and parent perspectives of sedation withdrawal assessment in critically ill children. Study 4 investigates how nurses use a sedation withdrawal tool, Studies 5 and 6 investigate what behavioural signs parents recognise and ascertain parents’ willingness to participate in withdrawal assessments. Key findings: Nurses found withdrawal behaviours difficult to interpret in critically ill children and there were differences in how these behaviours were construed. Parents identified a broader range of behaviours than included in existing tools. Most parents were eager to participate in the assessment. The elusive theoretical basis for the existing approach to withdrawal assessment may account for the lack of a standardisation and poor accuracy of the current tools. A model of the causal relationship between dependence and withdrawal is proposed. Recommendations: The model identifies the diagnostic criteria upon which a definition for Pediatric Withdrawal Syndrome may be based. These criteria also provide a novel framework for withdrawal assessment. Focussing on the shared diagnostic criteria and including the parent perspective of the child’s behaviours may aid the assessment and support decision-making.