Trial Design
The trial design is a parallel arm randomised placebo-controlled trial of an extended release formulation of MPH (OROS-MPH, Concerta XL) on ADHD symptoms, behaviour and functional outcomes in young male prisoners aged 16-25, meeting DSM-5 criteria for ADHD. Participants will be randomised to 8-weeks treatment with either OROS-MPH or placebo, titrated over 5 weeks to balance ADHD symptom improvement against side effects. 200 participants will be recruited with 1:1 ratio of drug to placebo. The duration of a participant’s follow-up is 8 weeks from the start date of the trial medication. Figure 1 illustrates the prisoner’s journey through the trial as the CONSORT diagram, which will be completed following the database lock. Figure 2 is a summary of all trial procedures and assessments.
Trial objectives
The primary objective is to establish the efficacy of OROS-MPH in reducing ADHD symptoms (inattention and hyperactivity-impulsivity) in young male offenders aged 16-25 who meet diagnostic criteria for DSM-5 ADHD.
Secondary objectives include: evaluating reductions in emotional dysregulation; the number of adjudications for antisocial behaviour and rule breaking in the previous 8-weeks; ratings of aggressive and/or disruptive behaviour by prison officers and education staff; attitudes towards violence; and self-report of well-being.
Additionally, we intend to investigate the hypothesis that improvements in secondary behavioural outcomes are mediated by improvements in ADHD symptoms or emotional dysregulation.
Study setting
Participants are recruited from two prisons. First HMYOI Isis in London (England, UK), a prison for sentenced young adults and category C offenders, defined as those that cannot be trusted in open conditions but who are unlikely to try to escape. Secondly, HMYOI Polmont in Falkirk (Scotland, UK), a holding facility for young offenders aged 16-21, with sentences ranging from 6-months to life. All participants were sentenced prisoners when screened for entry into the trial.
Eligibility criteria
Inclusion Criteria: Male, aged between 16 and 25 years (at consent for screening); English speaking (defined as sufficient to complete study assessments); able to provide informed consent (understand the information sheet and make an informed decision taking into account pros and cons of study participation); and meet clinical diagnostic criteria for DSM-5 ADHD.
The diagnostic criteria were defined as 5 or more symptoms of ADHD in either the inattentive or hyperactive-impulsive symptom domains, and 6 or more symptoms of ADHD in either the inattentive or hyperactive-impulsive symptom domains before the age of 12 years. Where it is not possible to gain enough clinical information to score childhood symptoms of ADHD, the operational criteria applied were adapted to include evidence of several ADHD symptoms with impairment starting before the age of 12 years, and 5 or more symptoms currently with moderate to severe impairment. In addition, persistent trait like (non-episodic) course of symptoms; impairments in two or more clinical or psychosocial domains and two or more settings from symptoms of ADHD; and onset of symptoms before the age of 12 years.
Exclusion Criteria: Lack capacity to give informed consent; moderate or severe learning disability, defined as IQ<60; serious risk of violence to the researcher; current major depression, psychosis, mania or hypomania; and past history of bipolar disorder or schizophrenia (exclude those with clear history of episodic mania/hypomania or psychosis unrelated to acute drug intoxication, but do not exclude on the basis of chronic emotional dysregulation i.e. irritability, frustration, anger or emotional-mood instability. Subjects were also excluded if they had medical contraindications to the use of stimulants (e.g. glaucoma, hypertension, cardiovascular disease or structural heart problem); were taking contraindicated medications during the 4 week prior to randomisation, showed drug seeking behaviour or craving (defined as drug seeking behaviour that is unusually severe and likely to affect the titration protocol due to unusual and excessive demands for drugs; or where there is a current withdrawal syndrome from an addiction disorder with drug dependency); receiving any ADHD medication between consent for screening and randomisation.
Trial medication:
OROS-MPH (Concert XL) is supplied as 18 mg capsules and placebo to match. Capsules are over-encapsulated and packaged in bottles of 46. Each bottle is assigned a unique randomisation number and the randomisation system allocates the right bottle to each patient. Over-encapsulation has been successfully adopted in previous studies to generate matched placebo to OROS-MPH. Piramal Healthcare UK Ltd. supply the investigational medicinal Product (IMP), placebo to match manufacture, clinical trials packaging, QP Certification and distribution for 200 patients. The Sponsor arranged the supply of Concerta 18mg tablets from the Marketing Authorisation holder, Janssen. Janssen provide the summary of product characteristics (SmPC), updated throughout the duration of the study.
The over-encapsulated active tablets are re-packed in HDPE bottles and take over the remaining shelf life of the study without the need for a stability program, as Concerta 18mg has a marketing authorisation for both HDPE and blister packaging. Placebo tablets are manufactured once. Trial medication over-encapsulation and packaging will be undertaken in 2 campaigns in order to accommodate a trial duration of up to 3.5 years. Concerta 18mg tablets typically has a maximum shelf-life of 3 years from the date of manufacture, however, by the time the product is repacked for the clinical trial, the remaining shelf life is likely to be under 2.5 years.
Over-encapsulation used ‘DBcaps’ capsules which are designed specifically for the blinding of clinical trial medication. We have to over-encapsulate the Concerta tablets with lactose capsule placebo, rather than make a matching placebo tablet, because Concerta tablets have printing on them and are of a distinct shape that would be difficult to manufacture and might infringe copyright. We sought advice on this from previous investigators using OROS-MPH and from companies who provide drug and placebo supplies for studies. Studies on the use of DBcaps have shown that encapsulation of tablets results in a lag time of 2–3 min in disintegration compared with the unencapsulated tablets. The pharmacokinetic properties of Concerta XL 18mg prolonged release capsules indicate release over several hours: following oral administration of Concerta XL to adults the drug overcoat dissolves, providing an initial maximum drug concentration at about 1 to 2 hours. The methylphenidate contained in the two internal drug layers is gradually released over the next several hours. Peak plasma concentrations are achieved at about 6 to 8 hours, after which plasma levels of methylphenidate gradually decrease.” (Section 5.2 of the SPC: https://www.medicines.org.uk/emc/medicine/8382).
Prescribing and titration procedures
Trial medication is delivered as prescribed daily, with participants observed to ensure they swallow the capsules. There is a daily record of compliance with the trial prescription. Both active medication and placebo are titrated in the same way. Treatment starts at an initial dose of 18 mg (1 tablet) for 1 week; and is then increased weekly over the following 4 weeks, in increments of 18 mg, to a maximum of 72 mg (4 tablets). Medication is reduced by 18 mg (1 tablet) if there is a limiting adverse event, in which case there will be no further increase in medication for the duration of the trial. Medication may be provided either once or twice daily up to the maximum daily dose. Titration upwards will be stopped if all 18 ADHD symptoms are scored as negligible (score of 0 or 1 on the CAARS) or absent. Unacceptable levels of adverse effects on the lowest dose of 18mg might lead to a cessation of treatment in a few cases.
A maximum dose of 72 mg was included for this trial because previous clinical trials indicated that a proportion of adults respond better at higher doses without unacceptable levels of adverse events; and because current licensing for Concerta XL up to 54 mg is based on dose levels for children and adolescents, rather than adults. NICE recommended a daily dose of MPH in adults to a maximum of 100 mg per day [7] and for Concerta XL the British National Formulary (No 62, September 2011) recommended doses up to a maximum of 108 mg in adults.
Strategies to improve adherence
We envisaged that adherence with allocated medication would present a challenge for around 20% of participants. Some offenders may not feel motivated to take the trial medication if they experience adverse effects or do not feel they are improving. They may also take medication intermittently because of the strict prison regime that allows for only a short time-window for leaving their cells to obtain medication from the medicine hatch on the prison wings. These cases are not expected to contribute to missing data. In our pilot study we accrued considerable experience in managing the expectation of offenders and providing the support needed to help participants adhere to the trial protocol. The following steps will be adopted to maximise adherence to medication:
- In the pilot, minor adverse effects (13%) were the most common reason for non-adherence to medication. This was linked to the observation that this population may be more sensitive to minor adverse effects, particularly changes in appetite, than community samples; perhaps reflecting the importance of meal times to prisoners. To maximise adherence to the protocol and minimise this as a potential source of missing data, we will take care to identify the early signs of minor adverse effects such as appetite loss and adjust the medication dose accordingly.
- Seven percent of the pilot sample did not wish to take medication in the mornings (08:00), which was the initial protocol followed in the pilot study. We then adjusted the protocol to allow for 12:00 medication for those that got up later in the day, worked mainly in the afternoons or had a strong preference for a 12:00 dosing, which resolved the problem. This flexibility in dosing time more accurately reflects dosing decisions in the community and provides a better match to patient’s daily routines.
- During the pilot study, prison staff did not always let patients out of their cells to receive medication or remind participants to get up on time. To resolve this problem, we initiated the use of research staff whenever possible, to assist in the delivery of medication by checking that prisoners were always out of their cells on time to receive trial medication.
- In the pilot study, treatment was disrupted for the Ramadan festival for several participants. We will take care to check that participants are not started on trial medication where religious customs might interfere with adherence to the trial protocol.
- In the pilot study, daily adherence to the trial medication reduced when participants were not reviewed weekly. One of the findings in the pilot study was the importance that prisoners gave to the weekly follow-up meetings when they can discuss their ADHD and response to the treatment process, in addition to completing study assessments. We will therefore offer weekly meetings with offenders throughout the 8-week trial.
- Nurse support in addition to a research assistant and medical staff will ensure that offenders are given the support they need to adhere to the protocol.
Concomitant treatments
Concomitant treatments are allowed with medications that are not contraindicated with MPH. All concomitant medications are recorded in the study database. Use of the following medications in the
4-weeks prior to the start of treatment with Concerta XL will lead to exclusion from the clinical trial, based on potential adverse drug interactions: clonidine, coumarins, monoamine oxidase inhibitors, moclobemide, and rasagline.
Baseline and outcome measures
The schedule of baseline and outcome measures, as well as the procedures for the trial, are listed in Figure 2. Baseline only measures are collected on all participants prior to randomisation and include descriptors of the study population and baseline moderators for further analysis of predictors of the treatment response. Primary and secondary outcome measures are collected at baseline prior to randomisation, and 8 weeks after the initial trial prescription. The primary outcome is the investigator rated Connors Adult ADHD Rating Scale (CAARS-O) at the 8-week outcome. The other 5-week and 8-week measures are secondary outcomes or mediator variables.
Investigator rated measures
- DIVA v2: Diagnostic Interview for DSM-IV ADHD [23]. DIVA 2.0. is semi-structured interview assessment used to capture the diagnostic symptoms and criteria for DSM-IV ADHD. The diagnostic algorithm applied to these data was modified for DSM-5 criteria.
- MINI 7.0.1: Diagnostic interview for comorbid mental health disorders [24]. MINI 7.0.1 is a semi-structured interview assessment used to capture DSM-IV diagnostic criteria for common mental health disorders. Sections completed included major depressive episode, suicidality, manic episode, hypomanic episode, panic disorder, agoraphobia, social anxiety disorder, obsessive-compulsive disorder, post-traumatic stress disorder, psychotic disorder and mood disorder with psychotic features, generalised anxiety disorder, and antisocial personality disorder. In addition to diagnostic categories evaluated at baseline only, we collected cross-disorder symptom checklist scores at baseline and the 8-week assessments.
- ZAN-BD: ZAN Borderline personality disorder [25]. A validated scale for the assessment of borderline personality disorder, used as a baseline moderator variable.
- CAARS-O: Conners Adult ADHD Rating Scale for ADHD symptoms [26]. The 8-week CAARS-O assessment is the is the primary outcome measure for this study. CAARS-O was also used as a secondary outcome at week 5, and to assist the psychiatrist in titrating participants onto the optimal trial medication dose. CAARS-O consists of the 18 DSM-IV ADHD symptoms, rated on a 4-point likert scale (0: not at all, never; 1: just a little, once in a while; 2: pretty much often; 3: very much, frequently). This scale and other closely similar scales have been extensively validated as outcome measures in previous clinical trials of adult ADHD.
- WRAADS: Emotional Dysregulation from the Wender-Reimherr Adult ADHD Diagnostic Scale [27]. We applied the emotional dysregulation items from an interview assessment of the WRAADS-ED, following previous publications on the treatment response of emotional symptoms in ADHD [20, 28].
- AES: Adverse events scale [29]. Scale of common adverse effects associated with stimulant medications for ADHD used with permission from the CADDRA website.
- CGI: Clinical Global Impression scale [30]. Scale used by the research psychiatrist to give an overall rating of clinical severity, and clinical impression of the clinical response and adverse effects of the trial medication.
Participant self-rating scales
Self-rating scales given to the participants for self-completion. The scale questions were usually read out to participants who gave their response accordingly.
- RPAQ: The Reactive-Proactive Aggression Questionnaire [31]. This scale is included as a baseline moderator capturing proactive and reactive forms of aggression.
- Weiss-CD: Weiss conduct disorder scale. This scale was included to capture conduct disorder symptoms as a baseline moderator.
- AUDIT-C and NIDA: Alcohol and substance abuse checklist using the AUDIT-C and NIDA quick screen to capture drug and alcohol use in year prior to the current prison sentence. The NIDA quick screen was adapted from the single-question screen for drug use in primary care by Saitz and colleagues [32]. Audit-C is validated as a quick screen for alcohol use [33].
- CTQ: Childhood trauma questionnaire [34], included as a potential moderator of the clinical response to MPH.
- B-ADHD: Barkley ADHD self-rating scale for DSM-IV ADHD symptoms [35] used as an initial screening instrument. Participants were considered to screen positive for ADHD if they had 4 or more symptoms scoring 2 or more in either the inattentive or hyperactive/impulsive symptom domain.
- ARI-S: The Affective Reactivity Index. [36]. A self-rating scale for irritability.
- MEWS: Mind Excessively Wandering Scale [37]. A self-rating scale that capture excessive spontaneous mind wandering, an aspect of psychopathology that is closely associated with ADHD and a strong predictor of ADHD associated impairment in daily life.
- BSI: Brief symptom inventory [38] is a self-rating scale that captures comorbid symptoms. Subscales include 9 symptom dimensions: somatisation, obsession‐compulsion, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism.
- MVQ: Maudsley Violence Questionnaire [39]. This scale was designed to capture beliefs associated with violence. The Machismo subscale relates to embarrassment over backing down and justification of violence in response to threat and attack; the Acceptance subscale includes the overt enjoyment and acceptance of violence in everyday life. In previous research the Machismo subscale showed the greater relationship to actual violence [39], and the greater reduction in our pilot study for this trial [14].
- CORE-M: CORE Outcome Measure [40]. This scale captures subjective well-being, problems and symptoms, life functioning, and risk and harm; designed to measure psychological distress before and after treatment.
Data from prison records and prison staff
Data will be collected from prison records and prison nursing and educational staff, relating to behaviour in the 8-weeks before the collection of the baseline measures. For cases of individuals new to custody presenting with significant behavioural problems, the retrospective baseline reporting period will be for a period of one month or more, to allow for initial behavioural problems that may arise when people first enter prison.
- Number of adjudications for antisocial behaviour and rule breaking (HMP Isis and HMP/YOI Polmont); and negative Incentive and Earned Privileges (IEPs) (HMP ISIS only)
- MOASP: Ratings of aggressive behaviour by prison staff using the Modified Overt Aggression Scale [41].
- BRCP: Ratings of behaviour by prison staff using behaviour report cards [42] by prison officers.
- MOASE: Ratings of aggressive behaviour by education staff using the MOAS. This item is optional, depending on whether prisoners are attending education sessions or not.
- BRCE: Classroom behaviour report card scored by education staff (HMP ISIS and HMP/YOI ISIS). This item is optional, depending on whether prisoners are attending education sessions or not.
- IEPs: Number of positive Incentives and Earned Privileges for positive engagement in education, occupational and rehabilitation programs (HMP ISIS only)
Baseline measures: The following measures are recorded at baseline. CAARS-O; WRAADS (3 subscores: temper, affective lability, emotional over-reactivity), weight, pulse, blood pressure, Weiss-CD, IQ (WASI), DIVA score and ADHD diagnosis, ZAN-BPD score, drug use in lifetime and alcohol use in the past year, RPAQ (2 subscales: reactive, proactive), MEWS, CTQ, MVQ, CORE-M, CGI, ARI-S, concomitant medications, BSI, MINI v7.0.1, and Adverse Events Scale.
Primary outcome measure: The primary endpoint is the level of ADHD symptoms measured on the investigator rated CAARS-O at 8 weeks post treatment initiation to address the question of efficacy of OROS-MPH on ADHD symptoms in young offenders meeting DSM-5 diagnostic criteria for ADHD. Investigator rated CAARS-O scores is a common outcome measure used in previous treatment trials of ADHD in the community; and measures the same list of 18 symptoms used as the primary outcome in nearly all other studies of Adult ADHD.
Secondary outcome measures: Secondary outcomes address important questions about the effects on comorbid symptoms and behavioural impairments that are commonly seen in offenders with ADHD. These are: critical incidents (adjudications) from prison records for the 8-week period (in two 4-week periods) from initiation of the trial medication to the 8- week assessments; ratings of aggressive behaviour by prison staff using the MOASP at 8 weeks; BRCP behaviour report cards from prison staff at 8 weeks; engagement with educational activities (including number of scheduled educational sessions, proportion of scheduled educational sessions attended, and reports of disruptive behaviour in education session reported at 8 weeks using the BRCE and MOASE completed by education staff only for those people involved in education); attitudes towards violence using the MVQ at 8 weeks, CORE-M at 8 weeks; g
Mediator measures: To address the secondary mediation hypotheses the following putative mediators are recorded at 5 weeks and at baseline: CAARS-O hyperactive/impulsivity and inattention sub-scores; WRAADS for emotional dysregulation. These measures are hypothesized to mediate treatment response in terms of secondary behavioural outcomes (critical incidents and prison staff classroom report cards). Critical incidents are taken from the prison records at 8 weeks and are recorded over the previous 4 weeks. The prison staff classroom report cards are recorded at 8 weeks and records behaviour over the preceding week.
Participant timeless
A flow chart of participant visits is illustrated in Figure 3.
Consent: There are two stages of consent. Initial consent 1 (screening and diagnostic step) allows for the use of screening questionnaires for ADHD, followed by a diagnostic assessment using the Diagnostic Interview for Adult ADHD (DIVA) interview for adult ADHD and review by a trained psychiatrist. During these pre-trial steps, patients who fail eligibility criteria will not be invited to continue and will not be asked to provide consent 2, to participate further in the trial (clinical trial step). The eligibility criterion which they are identified as failing will be noted. Individuals that do meet the diagnostic and eligibility criteria are invited to take part in the clinical trial, at which stage informed consent is requested to take part in the randomised controlled trial. Consent 2 is taken by the trial psychiatrists.
Patients complete baseline measures after informed consent to take part in the trial is agreed (consent 2). Once baseline assessments are complete, and eligibility checks completed and documented, the individuals will be randomised to one of the treatment arms.
Research visits: Following consent 1 for screening, and confirmation of the diagnosis of ADHD and eligibility by a psychiatrist trained in the assessment of ADHD, information sheets and consent forms for the controlled trial (consent 2) will be provided and discussed with potential participants (visit 1). Information sheets will be reviewed, and informed consent obtained for the clinical trial (visit 2). There is no limit on the time taken between visits 1 and 2 within the timeframe of the project. Potential participants will be encouraged to take as much time as they need to reach a fully informed decision about participation in the trial. Baseline data will be collected from participants, prison records and members of staff (visit 3). Once baseline data has been collected, and eligibility confirmed following medical review by a psychiatrist, participants will be randomised to treatment with placebo or OROS-MPH (visit 4). Trial prescriptions will be completed and given to the pharmacy. Medication should start within 1-week of Visit 4 (i.e. randomisation).
One week after the start of trial medication, the participants are reviewed, and trial medication titrated according to their clinical response and adverse effect profile (visit 5, week 1 titration). Symptoms of ADHD are measured using the CAARS-O, adverse events checked using the Adverse Events Scale, and blood pressure and pulse checked. This titration procedure is repeated at weeks 2 to 4 (visits 6-8). Five weeks after the start of medication (visit 9, week 5 assessment) the prescription (the titrated dose) is confirmed and maintained for the rest of the trial. At the week-5 assessment, outcome measures are completed by a research investigator for the CAARS-O, WRAADS and MEWS; and pulse, blood pressure, weight and adverse events scale. The final visit 10 is completed three weeks later after 8 weeks from the first prescription of the trial medication. At this visit all outcome measures are completed. As far as possible the information on clinical response derived during the titration visits (weeks 1-4) are not shared with other members of the research team; particularly with the investigator completing the week 5 and 8 outcome measures. Thus, potential unblindings based on the observed clinical response and adverse events will be minimised.
Sample size calculations
The total sample size to be randomised is 200.
The primary outcome is ADHD symptoms, measured using CAARS-O. The results of a single arm open label pilot study of young prisoners with ADHD who were given MPH showed a mean decrease of 25.0 points with a standard deviation of 9.1 [14]. This suggested a standardised effect size of SMD=2.75. It could reasonably be assumed that at least 20% of this effect might be attributed to the effects of MPH. On this basis, this study is powered to detect a standardised effect size of d=0.55. Assuming a standard deviation of 9.1, this would translate into a treatment difference of 5.0 points. This effect size is consistent with the results of a recent meta-regression analysis (29), which estimated the effect of treatment to be SMD=0.49 (95% CI 0.08, 0.64). The sample size calculation used G*Power version 3 and was based on the use of a t-test to compare the means of the treatment groups. In order to have 90% power at the 5% significance level to detect a standardised effect of SMD=0.55, this study would need to collect outcome data on 142 participants. Inflating for the expectation that loss to follow-up may be as high as 25%, a minimum of 190 participants should be recruited, with the target for the study set at 200.
25% loss is expected to be easily achievable since in the pilot 10% left the prison due to unexpected transfers from the prison and problems with adherence to trial medication was rarely followed by problems completing trial assessments.
Recruitment procedures
Participants will be recruited from HMYOI Isis (London) and HMYOI Polmont (Falkirk). Following consent to be screened for ADHD (consent 1), screening questionnaire data will be collected by the prison mental health teams using a DSM-IV ADHD symptom rating scale (25). Patients who screen positive will be invited to complete the Diagnostic Interview for Adult ADHD (DIVA) [27]. This will be followed by a clinical review by a psychiatrist trained in the diagnostic assessment of ADHD, including collateral information obtained from an informant whenever feasible. Following clinical review patients who meet diagnostic criteria for ADHD and meet the other eligibility criteria for the trial, will be invited to take part in the clinical trial.
Eligibility for the study will be further checked and recorded once the consent form (consent 2) has been signed and baseline assessments have been completed, prior to randomisation. Using an algorithm that applies the DSM-5 criteria to the DIVA interview data, cases will be checked to ensure they meet diagnostic criteria for DSM-5 ADHD. A clinical review by a psychiatrist trained in the diagnostic assessment of ADHD, will review all inclusion and exclusion criteria. The exclusion criteria of IQ less than 60 will be based on the 95% confidence interval for the IQ estimate from the WASI-II including IQ of 60, in combination with a clinical assessment by the psychiatrist to confirm that the participant has the ability to understand the rating scale and interview assessment questions, understand the information sheet and the study procedures and risks, and the ability to provide sufficiently detailed accounts of ADHD symptoms and behaviours, consistent with an IQ score greater than 60. Since there are no validated IQ tests for the visually impaired, including WASI-II, this criterion will be based on clinical judgement alone for participants with this impairment. This will also be the procedure for anyone unable to complete the WASI-II assessment due to severity of their ADHD symptoms or other mental health problems.
Withdrawal of Subjects
Participants have the right to withdraw from the study at any time for any reason, and healthcare staff have the right to withdraw patients from the trial if they consider the trial is having an adverse effect on the participants. However, where participants discontinue taking trial medication, we will invite them to remain in the study to complete trial assessments, thereby minimising loss of data. Should a participant decide to withdraw from the study, all efforts will be made to report the reason for withdrawal as thoroughly as possible.
Due to potential concerns about the interaction of trial medication with unknown psychoactive substances, if a participant discloses to any member of the research team that they have used “spice” i.e. synthetic cannabis or other unknown psychoactive substance, while participating in the study, a clinical evaluation will be made. If it is current use (defined as within the last two days) the study medication will be stopped. If it happened earlier in the study and this is considered an isolated incident the trial medication can continue. If the trial medication is stopped, the participant will remain in the study and will be asked to complete trial assessments. A clinical assessment will be made on a case by case basis as to the safety of restarting the trial medication after 48 hours from the time of stopping the trial medication.
Randomisation and allocation concealment
Randomisation to OROS-MPH or placebo will be at a 1:1 ratio. Randomisation is at the participant level and is performed using the King's Clinical Trials Unit’s (CTU) independent Randomisation Service, ensuring reliability and credibility in the randomisation process, with blinding of both investigators and participants. Randomisation is stratified by prison with variable block sizes to ensure that equal numbers of patients are allocated to the two arms within each prison stratum. Patient characteristics will not be considered in the randomisation process. However, we expect the drug treatment and placebo trial arms to be balanced in terms of cognitive ability, ADHD symptom severity and co-occurring psychosocial and mental health problems.
Prescriptions are completed by the trial psychiatrist. Each patient is allocated a kit (labelled carton) containing four labelled bottles, each containing 46 active or placebo tablets. Each kit and its bottles will be labelled according to Annex 13 guidelines and have its own randomisation/treatment pack number. The centralised randomisation system will allocate the correct treatment pack/kit to each patient during the trial.
Blinding
Blinding is maintained for all study investigators including the on-site researchers, pharmacy, statistical and data management teams. Investigators will be unblinded after the primary analysis is complete. The primary analysis dataset will not include any trial medication dosage data to ensure that the statistician remains blinded. We do however propose a sensitivity analysis to assess efficacy for those complying with tablets offered. This analysis will exclude those participants who took no trial medication on less than 75% of the days on which it was prescribed. Additionally, persons who withdraw from treatment or the trial or are released, transferred or deported will be excluded. We will not consider what proportion of the prescribed medication was taken on any given day.
We intend to use linear mixed modelling, which assumes that only variables included in the model predict missingness. We will assess empirically whether this particular missing at random (MAR) assumption is reasonable, using an independent statistician to maintain blinding if necessary. If the assumption is not reasonable, multiple imputation will be used instead to accommodate the missing data generating process and the statistician might need to become unblinded at this point, but investigators will remain blind until the primary analysis is complete. The Investigator must report all code breaks (with reason) as they occur on the case report form.
Emergency unblinding
Emergency unblinding will follow the standard operating procedures for the Kings Health Partners Clinical Trials Office. In circumstances where unblinding is deemed necessary, the first port of call will always be the local investigating team. Whenever possible the decision to unblind will be made the chief investigator, the principal investigator or clinically qualified staff working on the project. Out of hours, if clinically qualified members of the research team are not available, then the 24-hour emergency system will be used (ESMS). The ESMS system consists of a call centre which is manned around the clock by Information Scientists who have a minimum qualification of a life science degree to include toxicology or pharmacology. These Information Scientists are always available and are the direct line of communication to the number on the patient card. The Information Scientists will be trained in the specific details of this study and have direct access to one of the ESMS Consultant Physicians should clinical advice be required. Our Consultant Physicians practice general and internal medicine and specialise in clinical pharmacology and toxicology, ensuring clinical advice is available night and day. To maintain the overall quality and legitimacy of the clinical trial, code breaks will occur only in exceptional circumstances when knowledge of the actual treatment is absolutely essential for further management of the patient. The Investigator will always maintain the blind as far as possible.
Statistical analyses
A detailed statistical analysis plan will be developed by the trial statisticians in collaboration with the chief investigator and approved by the Trial Steering Committee before the trial database is locked. Analyses will be carried out by the trial statistician (RH) and checked by the senior statistician co-investigator (SL).
In the first instance data will be analysed under intention-to-treat assumptions, that is, participants will be analysed in the groups to which they were randomised irrespective of treatment received. Efficacy will be assessed by comparing primary and secondary outcomes between OROS-MPH and Placebo arms. The repeatedly measured continuous primary outcome (e.g. CAARS-O, WRAADS) will be analysed using linear mixed models. These models will contain outcomes from the last three (4, 5 and 8 weeks) post-randomisation time points as the dependent variables, where the primary outcome of interest is at 8 weeks. The trial arm, time dummy variables and interaction terms will be included as explanatory variables. Models will condition on baseline values of the outcome and the randomisation stratifier (prison). Random effects that vary at the participant level will be used to model the covariance structure between the repeated measures. The secondary count outcome at 8 weeks (number of critical incidents) will be compared between treatment arms using Poisson regressions to estimate incidence rate ratios (after conditioning on baseline counts and randomisation stratifiers). Logistic regression will be carried out for scheduled educational sessions attended. Parameters will be estimated using maximum likelihood.
Inferences will remain valid in the presence of missing data provided that the missing data generating mechanism is missing at random (MAR). More specifically this particularMAR assumptions stipulates that only variables included in the analysis model drive missingness. While we model several timepoints simultaneously, inferences will be made only at the timepoint of interest (8 weeks). Using linear mixed models means that we can allow variables measured and included in the model (e.g. previously observed values of the outcome including baseline values, trial arm, stratifier and post-treatment time point) to predict attrition and allows us to make use of all available data. We will also check empirically whether withdrawal from allocated treatments is predictive of missingness at 8 weeks. And if we found that such post-randomisation variables drive missingness we will consider using multiple imputation to accommodate such a MAR process. Mediation analysis using structural equation modelling will be used to partition the total treatment effect into mediated and non-mediated components.