PMC:3925439 / 9510-20337 JSON TXT

Annnotations TAB JSON ListView MergeView

2_test

{"project":"2_test","denotations":[{"id":"24524676-9700520-142841353","span":{"begin":3518,"end":3520},"obj":"9700520"},{"id":"24524676-11419452-142841354","span":{"begin":4523,"end":4525},"obj":"11419452"},{"id":"24524676-11419452-142841355","span":{"begin":4927,"end":4929},"obj":"11419452"},{"id":"24524676-9179098-142841356","span":{"begin":8923,"end":8925},"obj":"9179098"},{"id":"24524676-9179098-142841357","span":{"begin":9185,"end":9187},"obj":"9179098"},{"id":"24524676-9179098-142841358","span":{"begin":9668,"end":9670},"obj":"9179098"}],"text":"Method\n\nStudy design\nThe full protocol can be accessed at [http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3668233/].\nA summary of the protocol follows.\nEach child completed three pairs of one-week treatment periods (placebo versus stimulant medication), making a total of six weeks. To account for 'wash out’, no measure of efficacy was taken on the first two days of each seven-day period. The order of drugs in each cycle was determined by computer-generated random allocation, blinded to clinician, parent, teacher and patient. A computer generated randomisation schedule (random number sequence) created by the study statistician and held by the site pharmacies (and not accessible to investigators) predetermined the order of medication (stimulant (MPH or DEX) or placebo) in each cycle. A Research Assistant enroled participants, and the hospital clinical trials pharmacist sequentially assigned participants to random allocation sequences. Within each pair, the sequence was randomly allocated. For example, an actual sequence was (drug, placebo), (drug, placebo), (placebo, drug). At the end of the trial, the order of medications was unmasked, and compared with the parent and teachers’ observations of the child’s behaviour.\n\nOutcome measures\nThe investigators selected the following outcome measures:\n\nPrimary\n•Conners’ 3 Parent Rating Scales\n\nSecondary\n•Conners’ 3 Teacher Rating Scales\n•Behaviour Rating Inventory of Executive Function (BRIEF - Parent/Teacher/Self-Report)\n•Eyberg Child Behaviour Inventory (ECBI)\n\nParticipants\nParticipants in the study were purposefully sampled children who fitted the inclusion criteria and who were known to the Queensland Paediatric Rehabilitation Service, Brisbane. Australia (tertiary children’s public hospital).\n\nInclusion criteria\n1. Between 6 and 16 years old.\n2. Clinical diagnosis of moderate to severe traumatic brain injury, with severity based on duration of loss of consciousness, initial Glasgow Coma Scale (GCS) Score at presentation to the treating hospital, and duration of post-traumatic amnesia (PTA). Moderate TBI was defined as loss of consciousness for 30 to 60 minutes, or GCS 9 to 12, or PTA from one day to one week; and severe TBI was defined as loss of consciousness for greater than 60 minutes, or PTA greater than one week, or GCS less than 9.\n3. At least 12 months post injury.\n4. Clinically significant attention/concentration disorder or executive dysfunction including disorders of behavioural or emotional regulation that may respond to stimulants.\n5. At least two people (parent or other person and teacher) available to monitor the child’s symptoms.\n\nExclusion criteria\n1. Uncontrolled seizure disorder, moderate to severe hypertension, clinically significant anxiety, motor tics, Tourette syndrome, suspected or proven cardiac conduction problems, idiosyncratic reaction to sympathomimetic amines, history of drug abuse (including high caffeine beverages and appetite suppressants).\n2. Parents not able to fill out forms in English.\n3. Child’s school unwilling to participate.\n\nMeasures\nA. Baseline: age, gender, type and duration of symptoms, reason for referral, time post injury, presence or absence of ADHD (DSM-IV criteria) [22], demographics, and details of previous stimulant therapy.\nB. Weekly post randomisation: parents, teachers and capable children over 12 completed a weekly diary, including the following measures:\n1. Conners’ 3 Rating Scales for Parent and Teacher [23]. These are well-known valid and reliable scales [24] for measuring ADHD symptoms, in children aged 3 to 18 years of age. Subscales include inattention, hyperactivity/impulsivity, learning, executive functioning, aggression and peer relations, as well as subscales mapping onto DSM-IV criteria for ADHD (inattentive), ADHD (hyperactive-impulsive), ADHD combined, Conduct Disorder, and Oppositional-Defiant Disorder. Means and standard deviations for the global index vary dependent on age (for 6 year-olds the mean is 5.15 with a standard deviation of 3.97; means for 17 to 18 year-olds are 3.90 with a standard deviation of 4.00). The Global Index subscale was used for monitoring change in severity of behavioural symptoms over time. For teachers, the possible score ranges from 0 to 30. Raw scores are usually converted to T-scores or percentile scores relative to normative data. T-scores above 60 (percentiles above 84) are considered clinically significant.\n1. Behaviour Rating Inventory of Executive Function (BRIEF - Parent/Teacher/Self-Report); [25]. This assesses executive function behaviors of school-age children in the home and school environments (parent and teacher versions). The Global Executive Composite was analysed. The Global Executive Composite (GEC) is an overarching summary score that incorporates all of the BRIEF clinical scales. The possible range of scores for teachers is 0 to 24. It has satisfactory reliability and validity [25].\n1. Eyberg Child Behaviour Inventory (ECBI) [26]. This parent-rated scale measures two outcomes: Total Intensity scores and Total Problems scores. The intensity score is the total frequency of occurrence for the 36 behaviours and the problem score is the total number of behaviours for which the response is 'yes’. Both scales of the ECBI are continuous such that higher scores on the scale indicate a greater level of conduct-disordered behaviour and greater impact on the parent. There are two clinical cut-off scores: 127 on intensity (maximum = 252) and 11 on problem (maximum score = 36). Reliability and validity of this measure have been well-established [27,28].\nC. Post trial: post trial medication management plan.\n\nMedication\nPrior to the study, if not already stabilised on stimulant medication, the child was stabilised on an individualised oral dose of MPH or DEX. This was done to account for the wide variation in clinical dose required by children using stimulants and to identify significant side effects. If the child was not already stabilised on methylphenidate or dexamphetamine, prior to the trial commencement they were stabilised on an appropriate individualised dose. They were stable on this dose for approximately two weeks prior to commencement of the trial, unless already on stimulant medication prior to recruitment. The appropriate dose was one that provided a clinical improvement in target behaviours, and was well-tolerated with minimal side-effects, up to the maximal recommended dose stated by the product information. If unacceptable side effects occurred while taking one of the stimulants, the trial was offered using the other medication.\nChildren already on long-acting MPH and those judged likely to benefit from this were offered long-acting MPH trials at their clinician’s discretion. Other children were offered trials of short-acting MPH, or DEX, twice daily.\nStimulant and placebo were provided as identical opaque capsules, using encapsulation of either MPH or DEX powder, or over-encapsulation of the active whole tablet for long-acting MPH where the pharmacodynamics would not be altered by this. The pharmacy produced 3 × 1 week’s supply of each, packaged into Webster packs as per the computer generated randomisation order.\n\nProcedure\nThe University of Queensland Medical Research Ethics Committee and the Royal Children’s Hospital and Health Service District Human Research Ethics Committee, Queensland Health gave approval for this research study. Approval was obtained from Education Regulatory Authorities in Queensland (Queensland Government Department of Education and Training, and Catholic Education Archdiocese of Brisbane). Fully informed consent from parent and teacher (and assent for children over 12) was obtained.\nHospital clinicians determined subject eligibility, titrated stimulant medications and completed pre-trial (baseline) assessments. Parent/guardians provided consent forms, information sheets and questionnaires to schools for a teacher to complete weekly measures. Children then completed three pairs of one week treatment periods (placebo versus stimulant medication). All data were returned to research staff who provided a report to the treating doctor to discuss with the family.\n\nStatistical analysis\nThree types of Bayesian results will be presented here: (i) the mean of the posterior distribution of the mean difference between placebo and stimulant scores, which gives the best estimate of the overall effect size difference between treatments; (ii) the associated 95% credible region, which give intervals of uncertainty (in this case the 2.5 and 97.5 percentile) of the posterior distributions used in (i); and (iii) the posterior probability of the mean difference that stimulant scores were better than placebo scores, which describes the likelihood that the patients will favour the active treatment in future cycles. While seemingly analogous to confidence intervals in frequentist statistics [29], Bayesian credible regions have quite different interpretations and do not always coincide with confidence intervals as they incorporate problem-specific contextual information from the prior distribution and treat nuisance parameters radically differently [29]. However, like confidence intervals, a credible region that does not include the null value provides stronger evidence for the estimate in question than a credible region that includes the null value. Raw scores for the 18 cycles from the seven patients who completed at least one cycle were included in the statistical analyses.\nMeans, standard deviations (SDs) and ranges were calculated for stimulant and placebo periods. Replicating the technique described by Zucker et al. [29], hierarchical Bayesian random effects models were used to combine the n-of-1 studies to obtain estimates of treatment effectiveness for the group. Non-informative prior distributions were employed and inverse χ2 distributions (with the number of degrees of freedom given by the harmonic mean of the individual sample sizes) were used to give increased weight to within-patient heterogeneity estimates from patients with larger number of available measurements. Estimates of mean treatment difference, 95% credible regions (95% CRs), and posterior probabilities were determined. An important difference between treatment means was defined to occur if the 95% CR did not include the null value. Numerical results from the hierarchical Bayesian models were derived from computer simulation in WinBUGS [30]. Simulations of size N = 50,000 were run in five parallel chains after a burn-in period of 5,000 iterations. Convergence in the final samples was checked using visual plots of simulation histories and the modified Gelman-Rubin statistic [31].\nThe trial was to be stopped if the Data Safety Monitoring Board recommended stoppage due to safety concerns.\n"}

NEUROSES

PMC:3925439 / 9510-20337 JSONTXT

Annnotations TAB JSON ListView MergeView

2_test

NEUROSES

PMC:3925439 / 9510-20337 JSON TXT