Scientific background and rationale
Attention deficit hyperactivity disorder (ADHD) is a neurodevelopmental disorder that develops during childhood or early adolescence and frequently persists into adulthood. The disorder is defined by a persistent pattern of inattention and/or hyperactivity/impulsivity that interferes with or reduces the quality of functioning in daily life.1 The impairments associated with ADHD are often severe and are well documented. They include educational and occupational failure, transport accidents, difficulty forming social and personal relationships, and the development of behavioural problems, including antisocial behaviour and criminality.2–6 Many adults with ADHD struggle with mental health symptoms, such as a mental and physical restlessness, distractibility, emotional dysregulation, low self-esteem and sleep problems.3 A further source of impairment is the high rate of coexisting disorders associated with ADHD.7 These include general and specific learning difficulties, anxiety, depression, bipolar disorder, personality disorder and drug abuse disorders.
Attention deficit hyperactivity disorder affects around 5–7% of children.8,9 Prevalence estimates for ADHD in adults range from 2.5% to 3.4%,10,11 with the most recent review reporting an average prevalence of 2.8% for Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), ADHD.7 Currently, there are no prevalence estimates for ADHD using Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria. These estimates are expected to be slightly larger than those for DSM-IV because of changes in symptom thresholds for adults and the age at onset criteria. The population prevalence contrasts considerably with the disproportionately high rate of people meeting ADHD diagnostic criteria in prison populations, which averages around 26%.12 In this review and meta-analysis,12 studies using screening criteria for diagnosis had a significantly larger estimated ADHD prevalence (43%) than the subset of studies using a diagnostic clinical interview (26%), suggesting good sensitivity but a high rate of false positives when using screening tools. However, using clinical interview data,12 the rate of ADHD was estimated to be around 30% in male adolescents, 31% in male adults, 31% in female adolescents and 22% in female adults.
The National Institute for Health and Care Excellence (NICE) and the Scottish Intercollegiate Guidelines Network recommend stimulant medications, methylphenidate (MPH) or lisdexamfetamine, as first-line treatment for ADHD with significant impairment in children, adolescents and young adults.13–15 These medications target the core symptoms of inattention and hyperactivity/impulsivity, and may also lead to reduced levels of emotional dysregulation.16,17 Although psychoeducation and psychological interventions are also thought to be essential in the overall support provided for people with ADHD, they may not reduce core symptoms. Evidence for this conclusion comes from studies suggesting that psychological treatments may reduce coexisting problems (or enhance behaviours) relevant to the prison population, such as social skills, conduct problems and global mental health, but may not reduce core symptoms of the disorder.18–20
Despite the recommendations from national guidelines, there is continued uncertainty about the value of diagnosing and treating ADHD in offender populations21 and in adults more generally. Raman et al.22 reported that prevalence for medication use for ADHD averaged 0.03% in Western Europe (0.05% in the UK) among adults aged ≥ 19 years and 0.7% (0.64% in the UK) for children and adolescents. This is far below population prevalence for the disorder. The proportion of ADHD cases treated for ADHD in prisons is expected to be even smaller,21 yet clinical guidelines recommend the identification and treatment of ADHD at all ages, paying particular attention to the prison population, as well as those presenting with comorbid mental health disorders.14 One potential reason for the low rate of treatment of ADHD in prisons is the lack of data on treatment effects in adolescents and adults presenting with complex coexisting mental health and behavioural disorders, and the limited research of the prison population. Concerns are expressed that the common occurrence of mental health, neurodevelopmental and psychosocial problems might provide a better explanation for impulsive, overactive and inattentive states in young offenders, or might interfere with the treatment response in cases of ADHD. For these reasons, NICE13 recommended that drug treatment efficacy trials are needed in offender populations (section 5.18.1.413) and repeated this recommendation in 2013.23 The guideline13 stated that:
[T]here should be an assessment of efficacy in these groups (i.e. forensic and drug abuse populations) of the ADHD treatments already recommended for treatment in the community. Randomised controlled trial design is recommended.
There are two potential reasons why the response of ADHD symptoms to stimulant medications may be different for young adult offenders than for the samples of children, adolescents and adults involved in previous studies of ADHD in community settings. First, offenders present with an array of complex mental health problems that may better explain the states of inattentive, overactive, restless and impulsive behaviours used to define ADHD. This includes disorders commonly seen in offenders such as personality, anxiety, post-traumatic stress and substance abuse disorders. Second, nearly all previous trials of ADHD medications have been conducted in carefully selected samples with generally low levels of co-occurring psychosocial, behavioural and mental health problems. However, the co-occurrence of mental health disorders might modify the efficacy of drug treatments for ADHD.
One example is comorbid drug abuse. A meta-analysis24 of treatment trials found no effect of MPH on ADHD symptoms in ADHD cases with comorbid drug abuse [standardised mean difference (SMD) 0.08; p = 0.59], whereas there was a medium effect in samples without comorbid drug abuse (SMD 0.51; p < 0.00001). A review and meta-analysis25 of ADHD medications among patients with ADHD and drug dependency found only small reductions in the levels of ADHD symptoms following treatment (eight studies of MPH, three of atomoxetine, one of pemoline and one of bupropion), with an average SMD of around 0.30. When stratified by type of medication, significant reductions in ADHD symptoms were identified for MPH, with a mean dose of 62.2 mg per day across the eight MPH studies included in this review. Furthermore, an international consensus statement on treatment of ADHD with comorbid substance use disorder (SUD) concluded that several studies report lower effectiveness of standard-dose pharmacotherapy for the treatment of ADHD in patients with ADHD with comorbid SUD, and uptitration may be considered in patients who are unresponsive to standard doses.26 Finally, two recent studies of MPH found that a low MPH dose (i.e. < 36 mg) was associated with treatment discontinuation,27 and that ADHD patients with a SUD were, on average, prescribed 40% larger doses than those without a SUD after 2 years of treatment.28 These findings may be relevant to prison populations, for whom a history of drug abuse and ongoing access to drugs in prison is common.
Furthermore, concerns have been expressed about the potential for stimulant medications such as MPH to worsen coexisting conditions. The most recent research recommendations from NICE14 state that:
[N]o evidence was identified to justify different medication choices in people with ADHD and a history of psychosis, mania, or personality disorder. However, these groups are often excluded from trials. There are reasons (for example, mechanism of action of medication options, previous reports of adverse effects) to suspect that these groups may respond differently to different drugs, but a lack of trials to confirm this.
To address the questions raised, we have conducted a randomised controlled efficacy trial of osmotic releases oral system (OROS)-MPH, a sustained-release formulation of MPH, compared with placebo, in young male adult prisoners with ADHD who are aged 16–25 years.
Previous studies of methylphenidate in offenders with attention deficit hyperactivity disorder
This study followed on from an open-label study conducted by the chief investigator for this project that investigated the effects of OROS-MPH in 121 young male offenders in Her Majesty’s Prison and Young Offender Institution (HMP YOI) Isis in Southeast London, who met the criteria for DSM-5 ADHD.29 The current study follows similar procedures to the open-label trial. This included screening all prisoners aged between 18 and 25 years admitted to the prison using a DSM-IV symptom checklist,30 with diagnosis confirmed using the Diagnostic Interview for ADHD in Adults 2.0 (DIVA 2.0),31 followed by medical review from consultant psychiatrists trained in the diagnosis of ADHD. Prevalence of ADHD in the prison was estimated to be 19%, of which 78% met criteria for the combined type presentation of ADHD. Large pre–post treatment reductions with SMDs of > 2, unadjusted for multiple outcome measures, were seen for investigator-rated ADHD symptoms using the Conners’ Adult ADHD Rating Scale-Observer (CAARS-O).
Previous community studies demonstrate the efficacy and cost-effectiveness of MPH in reducing ADHD symptoms in children, adolescents and adults with ADHD.14 These conclusions are supported by a comprehensive review and meta-analysis of randomised controlled trials (RCTs) for MPH and other ADHD medications32 that, for MPH, found a large effect for children and adolescents [SMD –0.78, 95% confidence interval (CI) –0.93 to –0.62] and a moderate effect for adults (SMD –0.49, 95% CI –0.64 to –0.35). However, there are only limited trial data for the treatment of ADHD in young offenders presenting with a more complex mix of psychosocial, mental health and behavioural problems. One small RCT of MPH in a sample of 30 Swedish male prisoners with ADHD showed a very large effect (SMD 2.1) on ADHD symptom reduction.33 Although this study supports the treatment of ADHD in offenders, it may not generalise to the young adult offenders in the UK with mainly short-term sentences investigated in this study because of the small sample size, older age group, selection of severe ADHD cases with long-term sentences and treatment on a special prison unit.
Pharmacoepidemiological studies provide another important source of information, particularly for longer-term benefits of treating ADHD and the impact on a broader range of important outcomes relevant to offenders with ADHD.34 Unlike RCTs, these studies are representative of patients in real-world practice, include longer periods of follow-up and have large sample sizes. However, they are limited by confounding by indication, whereby patients may select treatments based on the nature or severity of their clinical condition. To account for this problem, these studies use within-individual designs, which compare periods of treatment with medication with periods of non-treatment.34 A good example of this approach is suicidal behaviour, which was found to be more prevalent among adults treated for ADHD with stimulants than among those who had not been treated. However, there was a lower rate of suicidality when periods of treatment were compared with periods of non-treatment in the same individual.35 These findings suggest that, although individuals with ADHD and suicidality may be more likely to be treated with MPH, during periods of treatment suicidality is likely to decrease. Using this approach, there is evidence for reductions in a range of outcomes relevant to offenders with ADHD, including serious transport accidents,36 depression,37 suicidality35 and SUDs.38,39 In patients with ADHD and comorbid bipolar disorder, there was reduced relapse of mania when patients were also treated with mood stabilisers, but there was an increased risk of manic relapse when not taking mood stabilisers.40 In patients with comorbid psychosis, there was no change in the risk of a psychotic relapse during periods of treatment and periods of non-treatment with MPH.41
Of specific relevance to this application was a pharmacoepidemiological study using Swedish national registry data of 25,656 male patients with ADHD, which found a sixfold-higher rate of criminal convictions over a 4-year period among ADHD patients than among controls.42 Regarding medication effects, they found a 32% reduction in the risk of criminal convictions using both ‘within’ and ‘between’ methods of analysis. Furthermore, these apparent protective effects on criminal convictions were seen for ADHD medications only (stimulants or atomoxetine), and not for commonly prescribed antidepressants, indicating the specificity of these findings to ADHD medications. Other important outcomes showing significant reductions during periods of treatment compared with periods of non-treatment included violent reoffending on release from prison43 and substance use.38,39
Another area of interest is the potential for improving educational and behavioural outcomes for offenders with ADHD.21 An initial descriptive study of ADHD in prisoners, conducted by co-applicant Susan Young, found a sixfold increase in critical incidents among prison inmates with high levels of ADHD symptoms, compared with inmates with low levels of ADHD symptoms.44 The association of ADHD with behavioural problems in the prison remained significant, even after controlling for antisocial personality disorder (ASPD), suggesting that this is an important outcome for clinical trials in prison populations. Another important outcome is symptoms of emotional dysregulation, such as irritability, anger and aggression, which are commonly seen in offender populations. Although emotional dysregulation is commonly seen in young adults with ADHD, it is a transdiagnostic set of symptoms seen across a wide range of other common mental health disorders.45–48 Meta-analyses of RCTs of ADHD medications found reductions in the symptoms of emotional dysregulation, including problems with temper control, mood lability and emotional over-reactivity.16,17 Hence, treatment of offenders with ADHD might also lead to significant reductions in emotional dysregulation and, potentially, aggressive or violent behaviour. The symptoms of ADHD are also known to interfere with education and employment because of a combination of restlessness, reduced attention span, forgetfulness and problems with planning and organisation.3 Treatment might therefore lead to greater positive engagement with prison education and rehabilitation programmes. In our open-label pilot study at HMP YOI Isis, we also found effects on all the secondary outcomes proposed for this study (all p < 0.001), including measures of emotional dysregulation, attitudes towards violence, the number of critical incidents and positive engagement with the education and rehabilitation programme.29
Potential benefits and risks of the trial intervention
For many years, ADHD was thought to be a childhood-onset disorder that had limited impact on adult mental health. However, the symptoms and impairments that define ADHD often persist into adulthood, with similar responses to drug treatments, such as OROS-MPH, to those seen in children and adolescents.3,32 Behavioural interventions can bring about further improvements in global clinical health, but only medication has been clearly shown to reduce the core symptoms of inattention and hyperactivity/impulsivity. For these reasons, clinical practice, expert consensus statements and national and international guidelines all indicate the need to diagnose and treat ADHD in adults using stimulants (MPH or lisdexamfetamine) as first-line treatments.14
Currently, several pharmacological treatments for ADHD are recommended for use in adults in the UK. Among the most used in clinical practice is an extended-release formulation of OROS-MPH, which is thought to have clinical effects lasting up to 12 hours from a single dose. OROS-MPH is currently licensed for use in children and adolescents, and the subset of adults with ADHD who were diagnosed as children; however, it is not licensed in Europe for use de novo in adults who were not diagnosed with ADHD as children. This has been a source of concern for clinicians, who feel that there is no rationale for excluding treatments just because a diagnostic assessment was not performed in childhood, generating a considerable unmet need,4,5 particularly in the prison population for those aged > 18 years, among whom the prevalence of undiagnosed ADHD is high.21,49 Furthermore, there is a risk that if ADHD is not included in adult mental health diagnostic assessments and treatment protocols, errors will be made, with patients put on antidepressants, antipsychotics or mood stabilisers for other conditions, while ADHD goes undiagnosed and untreated.
The potential value of treatment with OROS-MPH in young offenders with ADHD is that this is an established treatment with short- and long-term trial and post-marketing data to support the efficacy and safety of this medication in the treatment of ADHD. Potential benefits of treating young adult offenders with ADHD with OROS-MPH include improvements in clinical and behavioural outcomes. These include ADHD symptoms, emotional dysregulation, attitudes towards violence, critical incidents and engagement with educational and rehabilitation programmes.29 Demonstrating the efficacy and safety of OROS-MPH on ADHD symptoms and associated impairments may provide the data required to develop effective health-care pathways, including the use of OROS-MPH for a substantial group of young offenders. Establishing efficacy of OROS-MPH in this population would provide the foundation needed to establish long-term effectiveness studies with the potential for demonstrating significant reductions in criminal behaviour and improved health economic outcomes.
One potential risk that is often raised is the risk of abuse or diversion, particularly in a population of offenders with ADHD and high rates of substance abuse. Stimulants can be abused by crushing short-acting formulations, such as immediate release MPH, which can then be insufflated (snorted) or dissolved in water for injection, leading to a rapid entry of drug into the brain and the experience of euphoria. The importance of the route of administration was demonstrated in a study that showed that, although there is a dose–response relationship between intravenous MPH and a feeling of euphoria, this is not the case with oral MPH.50 This may not completely exclude the risk of abusing OROS-MPH in this way, but the evidence that this occurs is extremely limited; in addition, OROS-MPH is difficult to crush and it is difficult to extract MPH from OROS-MPH for injection. Risk of abuse was therefore reduced in this study by using this formulation of MPH. Furthermore, in the pilot study at HMP YOI Isis, we did not observe excessive drug-seeking behaviour for stimulant medication.29 Indeed, the young adult offenders being treated for ADHD were noted to be generally cautious about increasing the dose of medication and were titrated to modest daily doses, compared with community samples: 18% used 18 mg, 37% used 36 mg, 14% used 54 mg, 26% used 72 mg and only 4% used 90 mg.
Related to dosing, the other potential risks are the usual range of common adverse effects observed when treating ADHD with OROS-MPH. In the pilot open trial, the low doses the participants were titrated to was thought to be related to a greater focus by prisoners on minor adverse effects such as appetite loss, headaches and sleep problems, compared with patients in the community. This potentially could be reflected in lower dosing, reducing the potential effects on reducing the core symptoms of ADHD. Other potential risks on heart and blood pressure are routinely monitored. Diversion for non-medical use is another potential problem, but has been mainly identified as a risk in higher education settings to enhance education performance.51