Abstract
By his own admission, Mr Whitely, (2010) has conducted a sustained political campaign against the use of stimulant medications in Western Australia and further afield. Whitely makes the assumption of a positive outcome to his campaign, based on epidemiological and political arguments. It should be noted that political argument is by its nature adversarial and selective, so I would like to examine some of the assumptions and conclusions made by him in a number of contexts.
Epidemiology
First it should be pointed out that prescription rates are highly variable world-wide depending on demographic, socioeconomic and professional attitudes, availability and constraints (Hinshaw et al., 2011; Levy et al., 2010). A retrospective analysis of archival data on psychostimulant prescribing in South Australia indicated that while the rate of new prescriptions was found to be highly variable for both 1990-2000 and 2000-2006, age of initiation decreased and male-female ratio declined over both periods (Prosser and Reid, 2009), but the South Australian data showed periods of both marked increases and decreases in stimulant use. For example, a marked decrease across two major age groups (5-9 and 10-14 year olds between 1995 and 2000) was followed by an increase in 2004 to reach near 1995 levels in 2005, and then followed by an abrupt drop in 2005. Prosser and Reid (2009) point out idiosyncratic factors can result in significant changes in prescription rates. For example, in Western Australia during the period studied by them, 5 of 59 registered paediatricians accounted for 26% of prescriptions statewide. Changes in a small group such as relocations and retirements would result in significant effects.
Duration of treatment
In relation to duration of treatment, South Australian data shows the median duration of treatment was approximately 2.47 years from 1990 to 2000 while for the years 2001-2006 mean treatment length for girls and boys respectively was 1.98 and 2.05 years. US data shows similar mean durations and approximately 36% of children discontinued within 1 year (Bussing et al., 2005). Prosser and Reid (2009) concluded that the data on duration of medication suggests that significant numbers of children receive psychostimulant treatment for a relatively short period, and many children might be untreated during much of childhood and adolescence. In fact duration of treatment has received little attention and requires an understanding of when and why treatments are ceased. The epidemiology of treatment rates is more complex than might be apparent. For example, stimulants are often used relatively short-term to support remedial learning programmes.
Educational effects
Whitely also cites the Raine study of 131 children as showing a 950% probability of falling behind at school in Attention deficit hyperactivity disorder (ADHD) children with past stimulant use. While this is indeed a startling figure, it is unclear whether educational history and intellectual levels were compared.
Cardiovascular effects
Whitely cites Raine study findings of ‘significantly higher diastolic blood pressure’ in children who consistently received stimulant medication from 5-14 years. He equates this with a possible risk of permanent cardiovascular damage, while shorter term use is equated by Whitely with ‘risks of heart attacks and strokes’. However, little evidence is provided of these sequelae. In this regard, a retrospective cohort study from four US health plans was carried out on 1,200,438 children and young adults between the ages of 2 and 24 years (2,579,104 person-years of follow-up, including 373,667 person-years of current use of ADHD drugs (Cooper et al., 2011). The mean age of cohort members at baseline was 11.1 years and the mean length of follow-up for the cohort was 2.1 years for a total follow-up of 2,579,104 person years. Serious cardiovascular events such as sudden cardiac death, acute myocardial infarction and stroke from health plan data and vital records were recorded. Among 373,667 current users of stimulants, there were 7 confirmed events, compared with 25 confirmed events among 607,475 person-years of follow-up for former users and 49 confirmed events among 1,597,962 person years for non-users. The adjusted rate of serious cardiovascular events did not differ significantly among current and former users. The investigators found no evidence of increased risk for methylphenidate, while data was too limited for other drugs to fit regression models. Several alternative analyses produced similar results, but the upper limit of 95% confidence level indicated that a doubling of risk could not be ruled out, but the magnitude would still be low.
An associated study of young and middle aged adults compared current and new users of stimulant medications with non-use. An increased risk of serious cardiovascular events was not demonstrated (Habel et al., 2011). During 806,182 person-years of follow-up, the multivariable adjusted rate ratio (RR) of serious cardiovascular events for current use vs non-use of ADHD medications was 0.83 (95% CI). Among new users the adjusted RR was 0.77 (95% CI) and the RR for current use vs past use was 1.03 (95% CI). According to the authors, the upper limit of 1.28 corresponds to an additional 0.19 events per 1000 person-years at ages 25-44 years, and 0.77 events per 1000 person-years at 45-64 years.
While these studies are generally reassuring, a review of the relevant PubMed clinical literature on the efficacy and general tolerability of stimulant and non-stimulant ADHD medications in adults found that clinical trials of healthy adults with ADHD demonstrated increases in blood pressure (BP) (1-5mm) and pulse rate (4-10bpm) during treatment with immediate and sustained release stimulants (Hammerness et al., 2011). While not all of the increases were significantly different from placebo, it remained unclear whether there was a dose-response relationship. The authors also described complaints of palpitations, tachycardia and dyspnoea in up to 23% of stimulant treated individuals. They emphasize clinical recommendations to assess patients’ personal and family cardiac history prior to initiation of ADHD pharmacotherapy, and BP and heart rate monitoring where necessary, as well as the need for further research.
More recent reports (Kratochvil, 2012; Schelleman et al., 2011; Schelleman et al., 2012) found no statistically significant association of methylphenidate use with stroke or myocardial infarction. A 1.8 times increased risk of ventricular arrythmias or sudden death among methylphenidate users compared with nonusers was reported, but an inverse association with dose argued against a causal association, in the age group studied. Kratochvil points out that the massive resources devoted in the US to electronic health records should improve identification and understanding of rare but serious side effects.
Diversion of stimulants
Whitely also equates a decrease in amphetamine use among adolescents with a decrease in Western Australian prescription rates. It should be pointed out that dexamphetamine prescription rates have markedly decreased since the availability of slow release methylphenidate capsules, which are much more difficult to divert. Continuing reference to amphetamine diversion is thus outdated. A US systematic review of 21 studies, representing 113,104 subjects found different diversion patterns as a function of demographic features (sex, race, socioeconomic status, psychiatric status, medication preferred and route of administration) (Wilens et al., 2008). While the literature suggested that individuals with and without ADHD misuse stimulant medications, data derived from a US National Survey on Drug Use and Health found that the estimated number of lifetime illicit users was 4,524 plus/minus 210 with short acting methylphenidate, but was only 5 plus/minus 2 for extended release OROS MPH capsules (Concerta), though this estimate was limited by low numbers (Kroutil, 2006).
Conclusion
While responsibility for medication of children and adolescents is complex, unwarranted criticism can be damaging. I should point out that stimulants have been used in children with attention problems since 1937, and have stood the test of time. Stimulant actions are affected by age, developmental level, dose level and genetics and global conclusions may be misleading (Levy, submitted). Whitely argues for more rigorous external controls, but if these are based on uncritical assumptions they can do as much harm as good, and may deter those practitioners who actually work with children and families from utilizing evidence-based multimodal approaches to their work.