One year on from the Raine Study ADHD Medication Review – Will the analysis of this unique long term data source continue and if so can we trust those doing the analysis?

by Martin Whitely

In February 2010 a review of information from the Raine Study, a longitudinal study of the health and wellbeing of thousands of Western Australian children, provided the world’s first long term (8 year) data on the safety and efficacy of ADHD stimulants. It provided challenging evidence that amongst children diagnosed with ADHD those ‘medicated’ with stimulants had significantly worse outcomes than those ‘never medicated’. Specifically those ‘ADHD diagnosed and  medicated’ were 10.5 times more likely to be failing school than those ‘ADHD diagnosed and never medicated’. In addition the past use of stimulant medications was associated with permanently raised diastolic blood pressure.

Some of the authors of the ADHD data review (with a history of advocating the use of stimulants) were obviously expecting different results and tried to diminish the significance of its findings and revised their own methodology after results were in. Fortunately a robust committee process lead by a principled chairperson ensured the integrity of the review.

The Raine Study is a unique data source with the potential for further analysis. The ADHD medication review analysed the outcomes for the Raine Study children at age 14. A further six years of data (at ages 17 and  20) has now been collected. Whether this rich data source will be utilised, and who will do the analysis (and can they be trusted), are all questions that are yet to be answered.

There are literally thousands of short term, mostly pharmaceutical company sponsored studies on the effectiveness of ADHD stimulants. The majority of these studies conclude that ADHD stimulants are more effective than non-drug treatments in the short term. Aside from research bias, there are two reasons why these results are entirely predictable. Firstly, amphetamines (dexamphetamine) and near amphetamines like Ritalin alter behaviour immediately. And secondly, while the behaviour-altering effects of stimulants are almost universal (regardless of ADHD status), the effects of other forms of treatment are not. For example diet modification will be of little or no benefit if the underlying cause of behavioural problems is an undetected hearing problem.

The pharmaceutical companies exploit this by replicating short term studies and avoiding extended research that they realise may well demonstrate that in the long term amphetamines and near amphetamines are bad for growing minds and bodies. Ironically they justify this lack of long term research on ethical grounds, arguing either it is unethical to:1- experiment long term on children with pharmaceuticals or 2- deny children with a ‘neurobiological deficit’ (i.e. ADHD) access to ‘effective treatment’ (i.e. stimulants) by placing them on placebo or exposing them to unproven non-pharmaceutical treatments.

Until recently the best source of long term data on the safety and efficacy of stimulants was the three-year data from the Multimodal Treatment (MTA) Study which demonstrated no long term benefits and hinted at sustained harm (stunted growth and drug abuse). (for more information on the MTA refer to Pseudoscience Supporting ADHD)

However, published In February 2010 the Raine Study ADHD Medication Review provided the world’s first independent data on the long-term effects (eight years) of psychostimulant medication.^[1] The two most significant findings of the Raine Study ADHD Medication Review were:

1.   Long-term cardiovascular damage: ‘The most noteworthy finding in the study was the association between stimulant medication and diastolic blood pressure. Compared to not receiving medication, the consistent use of stimulant medication was associated with a significantly higher diastolic blood pressure (of over 10mmHg). This effect did not appear to be solely attributable to any short-term effects of stimulant medication, as when comparing groups who were currently receiving medication, it was found that those who had consistently received medication at all time points had a significantly higher mean diastolic blood pressure than those who had not consistently received medication in the past (difference of 7mmHg). These findings indicate there may be a lasting longer term effect of stimulant medication on diastolic blood pressure above and beyond the immediate short-term side effects.’ ^[2]

2.   School failure: ‘In children with ADHD, ever receiving stimulant medication was found to increase the odds of being identified as performing below age-level by a classroom teacher by a factor of 10.5 times.’^[3]

In addition the report indicated that there was a marginally negative outcome for both ADHD symptoms (inattention and hyperactivity) and depression with the long-term use of stimulant medication.^[4] (For more detailed information on the findings please refer to Excerpts from the Raine Study Data ADHD Long Term Drug Effect Review.)

The finding that amphetamine use may permanently raise diastolic blood pressure is of great significance. It had been previously recognised that while stimulants were in the patient’s system, heart rate and blood pressure were elevated, leading to the associated risks of heart attacks and strokes. But it was assumed that when the short-term stimulant effects wore off the cardiovascular system returned to normal.

The most startling finding was that past stimulant use increased the probability of an ADHD child falling behind at school by a massive 950 per cent. This finding completely undermines the hypothetical basis of medicating for ADHD. As stated in the review report the basis of the belief that amphetamines have long-term benefits are short-term studies, which ‘indicate that immediate management of ADHD symptoms allows children to function more effectively within a classroom. It is hypothesised that this makes children more available for learning and allows children to learn skills and concepts which are necessary to function well within a classroom in the future.’^[5] The analysis of the Raine Study data was the first time this hypothesis had been tested.

An advantage of using data from the Raine Study is that it reduces the risk of design bias as the original designers of the study had no idea the data would eventually be used to study ADHD.  In addition the body that commissioned the ADHD review, the Western Australian Ministerial Implementation Committee on ADHD (MICADHD), was an extremely diverse group. Opinions as to the safety and efficacy of stimulant medications within MICADHD were highly divergent.  This lack of consensus was a strength of the study as it limited the potential for ‘publication bias’ where there is a collective decision to bury results that are not in keeping with the consensus position of participants.

The suggestion that the Raine Study would be a possible source of long-term data on stimulant medication was first made by MICADHD members with a long history of prescribing and advocating the use of stimulants. They were obviously expecting very different results. I expected the results to show no long-term educational benefits or some adverse educational outcome from stimulants, but even I was surprised by the strength of the negative outcome. Initially, the medication proponents on MICADHD tried to claim that the outcomes for the medicated children were most probably worse than those for un-medicated children, because the medicated children had more severe ADHD. However, as a member of the MICADHD committee I insisted on a comparison of the groups at age five, which was prior to any of the children having been medicated. This analysis established that there were no statistically significant differences in developmental, behavioural and health measures before the children were medicated.

This did not prevent subsequent efforts to explain away this research using the unsupported hypothesis that differences in educational performance and the increase in blood pressure were due to pre-existing differences. Professor Ian Hickie from the Brain and Mind Research Institute dismissed the poor educational outcomes saying, ‘typically those kids who go on the medication are considerably worse to start with’.^[6] Hickie’s comment needs to be seen in the light of his commercial ties to pharmaceutical companies. His CV acknowledges receipt of $411,000 from four different pharmaceutical companies, mostly for research on psychotropic medications for depression, psychosis and bipolar disorder.^[7] Even supposing Hickie’s criticisms had been valid, and the children who had been medicated ‘were considerably worse to start with’, if amphetamines worked, they should have been performing at least as well at school as those with moderate, un-medicated ADHD.

As with all studies there are limitations with the ADHD medication review. While the sample size (131) was small, ‘it was larger than those in many short-term studies that supported the use of stimulants as a safe and effective treatment for children with ADHD’.^[8] Although the evidence now available from the Review does not prove that amphetamines cause failure at school and permanent cardiovascular damage, it is significant because there is so little other long term research to guide clinical practice.

The lack of enthusiasm for the findings of the research demonstrated by some of its authors diminished its impact when it was first published in February 2010. If not for the robustness of MICADHD Committee processes and the integrity of its chairperson, Professor Lou Landau, it challenging findings may have never been published. However, the MICADHD Committee is now dissolved. The question of if, and just as importantly who, will be conducting any future ADHD medication analysis of new Raine Study data (at age 17 and 20) is unclear.

These are important questions as the Raine Study is a unique and significant data source. The process of further ADHD medication analysis of Raine Study data must be transparent and independent of undue influence. Vigilance is required.

Appendix

What is the Raine Study and how is the data used to analyse the safety and effectiveness of ADHD stimulants?

‘The Western Australian Pregnancy Birth Cohort (Raine) Study is an ongoing longitudinal study following 2,868 children. The study began in 1989 as a pregnancy cohort of women enrolled at or before the 18th week of gestation from the public antenatal clinic at the principal obstetric hospital in Perth, Western Australia or nearby private practices. Since 1989, data has been collected from the participants (both the mother and her child) at regular intervals including when the child turned 1, 2, 3, 5, 8, 10, 14 and 17 years old.’^[9]

Across the globe there are other similar long term, large scale, studies into children’s health and wellbeing, however, the Raine Study is the only one in a location which has had high rates of ADHD stimulant child prescribing. This unique potential data source was recognised by the Western Australian Ministerial Committee on ADHD (MICADHD) who commissioned a review of the Raine Study data in response to a 2004 Western Australian Parliamentary Inquiry that recommended “research into the safety and efficacy of the long run use of psychostimulant medication”.^[10] ,^[11]

By age fourteen ‘of the 1785 adolescents in the sample, 131 (7.3%) had received a diagnosis of ADHD.’^[12] At age five none of the 131 had taken ADHD stimulants. The comparison of the groups at age five showed there were no statistically significant differences in symptom severity or health measures. The statistically significant differences that existed at age 14 occurred between age five and fourteen, after some of the children were medicated. To the extent that (non statistically significant differences) existed at age 5 these were ‘controlled for by using the ‘propensity for medication’ score, the symptom severity before commencement of medication treatment, and a number of sociodemographic measures.’^[13]

The data for the 131 ADHD diagnosed children was grouped for analysis in two different ways:

The first was by ‘Current Use of Medication’. At age fourteen, “21 (16.0%) were using medication… and had used it consistently since being diagnosed with ADHD; 40 (30.5%) were using medication.. but had not consistently used it in the past; 41 (31.3%) had used medication in the past be were not using it at age 14, and; 29 (22.1%) had not reported using stimulant medication.”^[14]

The second was by ‘Ever Use of Medication’. Of the 131 children, “21 (16.0%) reported the use of stimulant medication at all three follow-up points (8, 10, 14 years), 42 (32.1%) at two follow-up points, and 39 (29.8%) at one of the follow-up points. 29 (22.1%) reported no use of stimulant medication at any of the follow-up points.”^[15]

These two measures gave the capacity to analyse the differences attributable to differences resulting from 1- residual effects of medication, 2- current effects of medication, 3- effects of extended duration of medication.

http://www.health.wa.gov.au/publications/documents/MICADHD_Raine_ADHD_Study_report_022010.pdf

[4] ibid., p. 5

http://www.health.wa.gov.au/publications/documents/MICADHD_Raine_ADHD_Study_report_022010.pdf

[5] ibid., p. 30

http://www.health.wa.gov.au/publications/documents/MICADHD_Raine_ADHD_Study_report_022010.pdf

; The short term studies referred to in the Raine Study are Howard B. Abikoff, et al., ‘Methylphenidate effects on Functional Outcomes in the Preschoolers with Attention-Deficit/Hyperactivity Disorder Treatment Study (PATS)’, Journal of Child and Adolescent Psychopharmacology, 17(5), 2007, pp. 581–92; C. L. Carlson & M. R. Bunner, ‘Effects of Methylphenidate on the Academic Performance of Children with Attention-Deficit Hyperactivity Disorder and Learning Disabilities’, School Psychology Review, 22(2), 1993, pp. 184–98; Irene M. Loe & Heidi M. Feldman, ‘Academic and educational outcomes
of children with ADHD’, Journal of Pediatric Psychology, 32(6), 2007, pp. 643–54.

[6] Interview with Professor Ian Hickie on ABC PM program with Mark Colvin, ‘New Research Reignites Debate over ADHD’, 17 February 2002. Available at http://www.abc.net.au/pm/content/2010/s2822748.htm (accessed 15 March 2010).

[7] Professor Ian Hickie received the following grants totalling $411,000 from pharmaceutical companies: $10,000 from Roche Pharmaceuticals (1992); $30,000 from Bristol-Myers Squibb (1997); $40,000 from Bristol-Myers Squibb (1998–1999); $250,000 from Pfizer Australia (2009); $81,000 from Pfizer Australia (n.d.). Cited in Ian Hickie, Curriculum Vitae, last updated 23 August 2009.

[8] Government of Western Australia, Department of Health, Study raises questions about long-term effect of ADHD medication, Media Release, 17 February 2010.

[9]  Draft: Long-term outcomes associated with stimulant medication in the treatment of ADHD in children, Perth, Telethon Institute for Child Health Research (January 2010): p14

[10] Education and Health Standing Committee: Annual Report 2003-2004, Legislative Assembly, Parliament of Western Australia, Perth, October 2004 recommendation 2, page 42

[11] Western Australian Ministerial Implementation Committee for Attention Deficit Hyperactivity Disorder, Raine Attention Deficit Hyperactivity Disorder Study, September 2009

[12] Draft: Long-term outcomes associated with stimulant medication in the treatment of ADHD in children, Perth, Telethon Institute for Child Health Research (January 2010): p25

[13] Draft: Long-term outcomes associated with stimulant medication in the treatment of ADHD in children, Perth, Telethon Institute for Child Health Research (January 2010): p56

[14] Draft: Long-term outcomes associated with stimulant medication in the treatment of ADHD in children, Perth, Telethon Institute for Child Health Research (January 2010): p26.

[15] Draft: Long-term outcomes associated with stimulant medication in the treatment of ADHD in children, Perth, Telethon Institute for Child Health Research (January 2010): p26