Dose of Methylphenidate during Service Transition for Adults with ADHD

Abstract

Objectives:

We wanted to investigate how one element of the transitional process of adolescents to an adult ADHD service, namely the use of medication, fared when compared to the recommendations of national guidelines.

Methods:

We did a chart review of the dose of stimulants in a cohort of transitional patients after they were transferred to an adult ADHD service, whilst investigating if other variables such as severity of ADHD, age, gender and comorbidity played any role in determining the dose of stimulants at transition.

Results:

The dose of stimulants when calculated in mg/kg was almost half the recommended whist the patients were also severely symptomatic. Reported comorbidity with Autistic Spectrum Disorders was also very high.

Conclusions:

A handover approach of adolescents with ADHD to an adult service, may hide gaps at least as far as prescribing is concerned. This gap may need considerable resources to address successfully. We suggest that establishing transitional processes may help minimise this problem.

Introduction

Attention deficit hyperactivity disorder (ADHD) is defined in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) [American Psychiatric Association, 1994] as a clinical syndrome presenting with ‘high levels of inattentive and/or hyperactive and impulsive behaviours in early childhood that persist over time, pervade across situations, and lead to notable impairments’. Others [Rowland et al. 2002] define ADHD as the most common neurodevelopmental disorder of childhood and, as such, ADHD is a common disorder. In the UK, a survey of 10,438 children between the ages of 5 and 15 found that 3.62% of boys and 0.85% of girls had ADHD [Ford et al. 2003]. In adults, the prevalence is reported to be from 2.9% [Faraone and Biederman, 2005] to 4.4% [Kessler et al. 2006].

For ADHD, it is the case that patients will be transferred from children's health services to adult services around the age of 18. The National Institute for Clinical Excellence (NICE), went as far as recommending models of how this should take place [National Institute for Health and Clinical Excellence, 2008a] and the concept of a transitional process [Department of Health, 2006] has been well articulated by policy makers. However, despite the belief that the policy initiatives outlined above are moving on well [Department of Health, 2008], the fact on the ground for ADHD is that they are not moving well at all [Singh, 2009]. The main reason for this is the not only the general lack of services for adults with ADHD in the UK [Tettenborn et al. 2008; Verity and Coates, 2007], but the disparity among providers on the use of transitional protocols for adults with ADHD [Singh et al. 2008].

As a consequence of this disparity, apart from the potentially negative effects on the clinical care of adults with ADHD, there is the lack of UK relevant transitional data to inform service development. This paper aims to begin to bridge that gap by providing information on one of the aims of the transition process which is the review of medication.

Method

All patients in transition from adolescence to adulthood referred from their general practitioner, paediatrician or child and adolescent psychiatrist to the Yorkshire service for adults with ADHD between June 2009 and February 2010 were included. The transitional protocol was not yet in place during this time and therefore the process of transfer was of hand-over. To meet the inclusion criteria, all of these patients were taking medication (methylphenidate) for the management of their ADHD symptoms at the time of their first consultation. Twenty two patients met the inclusion criteria.

Of these patients, we collected data retrospectively and compiled a spreadsheet containing the following information: age, gender, weight, medication (including preparation) comorbidity with autistic spectrum disorders (ASD) as reported by referrer, comorbidity total (other mental health disorders diagnosed at assessment plus ASD as reported by referrer), and dose of medication on first consultation. From this matrix, the dose of medication in milligrams per kilogram was calculated and methylphenidate dose equivalents as described by NICE [National Institute for Health and Clinical Excellence, 2008a] were used. We did not make a specialist ASD assessment to confirm or disprove this diagnosis.

The weight in kilograms of each patient was checked during the first consultation as well as other physical parameters such as blood pressure (BP), pulse and height. At the same consultation, ADHD symptom severity was assessed using the investigator-administered 18-item total ADHD symptom score which is the sum of the inattentive and hyperactivity/impulsivity subscales from the Conners' Adult ADHD Rating Scales (CAARS) and has a maximum score of 54 [Conners et al. 1999]. The medication (including preparation) and current prescribed dose were also confirmed. The first set of analyses examined whether there were any factors associated with either the Conners' score or the dose given. An examination of the distribution of the Conners' score suggested that this was approximately normally distributed. As a result, the unpaired t-test was used to compare this measure between patients with and without comorbidity. There was insufficient data to formally compare between genders.

Additional analyses compared the dose between patients with and without comorbidity and also between genders. The dose values were found to have a positively skewed distribution, and were not normally distributed. Therefore, the Mann—Whitney U-test was used for these analyses. Owing to the distribution of the values, the median was used as the summary measure in preference to the mean.

The final analysis of this data examined the association between age and dose or gender and also between Conners' score and age. These associations were examined using Pearson correlation.

Results

Of the 22 patients, only one was female and the mean age was 19.7 (SD = 1.93) years old. The mean Conners' score was 30.1 (SD = 12.8) and the mean dose of stimulant (mg/kg) was 0.56 (SD = 0.30). The total comorbidity including ASD and other mental health disorders was 31.8% whilst the reported comorbidity by referrer with ASD was 27.3%.

The first set of analyses examined the relationships between variables in the dosing dataset and the results are presented in Table 1. The results suggested that there was no difference in Conners' score between patients with and without any comorbidity. The figures presented are the mean and standard deviation score in each group, and also a p-value indicating the significance of the score.

Table 1.

Relationships between variables and dose.

Variable	Category	Conners' score mean (SD)	p-value
Comorbidity with ASD	Yes	29.3 (20.6)	0.91
	No	30.3 (11.6)
Total comorbiditya	Yes	32.7 (15.3)	0.55
	No	28.4 (11.7)
Gender	Male	30.1 (13.4)	–b
	Female	30.0 (0.0)

^aTotal comorbidity is comorbidity with other mental health disorders and does not include ASD.

^bNot possible to perform an analysis, as only 1 female with data on Connors score.

ASD, autistic spectrum disorders.

Using the Mann—Whitney U-test to compare the dose of stimulant (mg/kg) given between groups, there was found to be no difference in dose between groups for any of the three variables examined. The figures reported in Table 2 are the median dose (mg/kg) in each group, and also the interquartile range, the interval containing the middle half of the data. p-values indicating the significance of the results are also given.

Table 2.

Median dose of stimulants per variable.

Variable	Category	Dose median (IQR) mg/kg	p-value
Comorbidity	Yes	0.46 (0.36, 0.70)	0.83
	No	0.51 (0.31, 0.75)
Total Comorbidity	Yes	0.45 (0.36, 0.70)	0.87
	No	0.60 (0.28, 0.73)
Gender	Male	0.47 (0.36, 0.70)	0.91
	Female	0.53 (0.31, 0.75)

IQR, interquartile range.

The next set of analyses used Pearson correlation to examine the association between the variables measured on a continuous scale, with the results summarized in Table 3. The results indicated no evidence of an association between the Conners' score and either dose or age. The figures are the correlation coefficients and their associated p-values indicating the significance of the results.

Table 3.

Effects of dose or age on Conners' score.

Variable 1	Variable 2	Correlation coefficient	p-value
Conners' score	Dose	0.11	0.70
Conners' score	Age	−0.15	0.60
Dose	Age	0.42	0.05

However, there was some evidence of an association between age and dose, although this result was only of borderline statistical significance (p = 0.05) suggesting that the dose increases with increasing age.

Discussion

The finding that the dose of methylphenidate being taken by patients with ADHD was approximately half of what is normally recommended for that age group [Biederman et al. 2010] at the time of handover needs to be explained. This discrepancy in dose occurred despite most patients being severely symptomatic as assessed by the 18-item total ADHD symptom score of the CAARS during the initial consultation. The British National Formulary [Joint Formulary Committee, 2010] states that for children aged between 6 and 18, 2.1 mg/kg daily of methylphenidate in two or three divided doses can be used, up to a maximum of 90 mg daily. Although these dose ranges are outside of the UK marketing authorisation for all methylphenidate preparations, they are recommended by NICE but were not used in this sample population.

When a thorough investigation of the types of services provided to US ADHD children were studied and analysed during the National Ambulatory Medical Care Survey (NAMCS) in 1989 for a 7-year period, it was recognized that patterns of services for children with ADHD were changing. During that period, there was a 2.9-fold increase (p < 0.05) in the population-adjusted rate of ADHD patients prescribed stimulant pharmacotherapy and a 2.6-fold increase (p < 0.05) in the population-adjusted rate of ADHD patients prescribed methylphenidate [Robison et al. 1999]. This increase could be attributed to a doubling in the rate of diagnosing ADHD from less than 1% in 1989 (0.74%) to almost 2% (1.9%) in 1996 with this trend yet to be reversed possibly due to underdiagnosis [Kewley and Orford, 1998]. As a result of this increase, services were strained and major gaps between research base and clinical practice were identified with only 50% of children receiving care that corresponded to guidelines of the American Academy of Child and Adolescent Psychiatry [Hoagwood et al. 2000].

ADHD services around the UK are disparate [Tettenborn et al. 2008] and we would suggest that if audited against the recent NICE Guideline [National Institute for Health and Clinical Excellence, 2008b], the gaps between research base and clinical practice would be at least as high as in the US. Guidelines for ADHD existed in the UK from the beginning of the millennium [Nutt et al. 2007; National Institute for Health and Clinical Excellence, 2006, 2000; Scottish Intercollegiate Guidelines Network, 2001; British Psychological Society, 2000] and although the complete NICE Guidelines only arrived late in 2008 [National Institute for Health and Clinical Excellence, 2008a], there was adequate time for implementation. Indeed, the 20 centres of excellence participating in a multinational study appeared to broadly follow the recommendations set out in national guidelines at the time [Tettenborn et al. 2008], but the same may not translate elsewhere in the country. This disparity in service delivery, in our opinion mainly a result of underinvestment, could have been the reason behind the disparity in the doses of stimulants for the sample we collected. Furthermore, the lack of clinical pharmacy services to the children and adolescent mental health and community paediatric teams may be a contributing factor as to why guidelines referring to medicines management are not implemented and/or adhered to.

As far as transition is concerned, the experience of UK community paediatricians is that there are not many places that their patients can go when they reach adulthood [Marcer et al. 2008]. The suggestion that ADHD is likely to become increasingly important for primary care [Thapar and Thapar, 2002] and that generic teams may take over the care of adults with ADHD cannot be supported not only by our findings, but by the fact that many families of children or adults with ADHD have complex ongoing needs which merit specialist input [Salmon and Kemp, 2002]. The transition period is therefore a landmark process where not only people come together, but also different service cultures and therapeutic approaches aiming to meeting the patients' needs. We would expect that during this process and with clinical pharmacy input, disparities in medicines management would be addressed.

Our data also showed that comorbid disorders are common in adults with ADHD. Anxiety disorders, substance abuse disorders and mood disorders are all highly prevalent comorbidities in this patient population, and there is also a significant incidence of antisocial disorder [McGough et al. 2005; Biederman et al. 1993; Shekim et al. 1990]. We were not attempting to study this area with this small sample and maybe that is why we found that patients with or without comorbidity had similar Conner's results and were prescribed similar doses of medication.

This study has limitations: it relies on a small predominantly male sample from a pool covering approximately 1.2% of the national population and thus may not be immediately generalizable. However, it is a rare study in that it demonstrates the need for a transitional process from the general adult psychiatry side and does this by identifying that reviewing medication during this process may benefit the patient.

Conclusions

With an increased trend in diagnosis and prescribing for children with ADHD, adult services will find themselves under more pressure to receive this population group. NICE has offered some suggestions on how these arrangements should take place but in our opinion, has not stressed enough how essential the transitional process is. By not having a clear partnership in the transfer of care for these patients to adult services, more resources may have to be devoted at a later stage to address important gaps in care such as in medicines management.