Advanced Economy Outlier: Even in China's largest cities, ADHD is seldom treated with pharmaceuticals

The Background: 

Randomized clinical trials have shown ADHD medications are effective in reducing core ADHD symptoms. Moreover, large observational studies indicate that these medications are associated with lower risks of real-world outcomes, including injuries, crime, transport crashes, suicide attempts, and unnatural-cause mortality. 

Sweden’s ADHD medication use has soared. From 2006 to 2020, children’s use rose almost fivefold, and adults' use more than tenfold. This places Sweden among the highest globally in ADHD prescriptions. 

Research indicates that rising prescription rates are due to changes in diagnostic criteria and their interpretation, parental perception, and greater awareness of ADHD, rather than an actual increase in its prevalence. 

Sweden has a single-payer health insurance system that covers virtually its entire population, as well as a system of national registers that link health care records to other population databases.  

The Study:

A research team based in Sweden used that data to explore how the impact of ADHD medication on self-harm, injuries, traffic crashes, and crime has evolved with the dramatic increase in ADHD prescription rates. The team hypothesized that effects would decrease as medications were prescribed to a broader group of patients, including those with fewer impairments and risky behaviors who might not derive as much benefit from pharmacotherapy. 

The team identified all individuals aged 4 to 64 who were prescribed ADHD medication and living in Sweden in the fifteen years from 2006 through 2020. From this base cohort, they selected four specific cohorts for self-harm, unintentional injury, traffic crashes, and crime, consisting of individuals who experienced at least one relevant event during the study period. 

They used a self-controlled case series (SCCS) design to explore the link between ADHD medication use and outcomes. This approach allows individuals to serve as their own controls, accounting for confounders like genetics, socioeconomic status, or other constant characteristics during follow-up. 

A non-treatment period was defined as a gap of 30 days or more between two consecutive treatment periods. To examine the link between ADHD medication use and outcomes, the team divided follow-up time into consecutive periods for each individual. A new period began after a treatment switch. They estimated incidence rate ratios (IRRs) to compare the outcome event rates during medicated periods with non-medicated periods for the same individual. 

The team examined how ADHD medication outcomes varied with prescription prevalence across three periods: 2006-2010, 2011-2015, and 2016-2020, during which ADHD medication use continuously increased. 

The overall cohort encompassed almost a quarter million ADHD medication users: just over 57,000 for 2006-2010, just over 127,000 for 2011-2015, and slightly over 200,000 for 2016-2020. 

The Results:

ADHD medication use was linked to significantly lower rates of all studied outcomes during the study period. However, as prescription rates increased five to tenfold in the population, the strongest association for reduction in self-harm was observed between 2006 to 2010 (23% reduction in incidence rate) and was slightly reduced (below 20%) in the two more recent periods, though this change was not statistically significant.  

On the other hand, there was a significant decreasing trend in the reduction of incidence rate ratios for unintentional injury, with a 13% reduction in incidence rate in 2006-2010 decreasing over the two more recent periods to half that amount, 7%. For traffic crashes, a 29% reduction in incidence rate significantly diminished by more than half, to 13%. For crime, a 27% reduction in incidence rate from medication use significantly declined to 16%. 

When considering methylphenidate prescriptions only, these effects were partially attenuated for crime. A 28% reduction in the incidence rate for crime in 2006-2010 dropped to 19% in the two most recent periods, but the trend was not statistically significant. Nevertheless, there were no significant differences from the results in the larger cohort in any of the other categories.   

The Interpretation:

These outcomes were consistent with the team’s hypothesis. The researchers concluded, “While ADHD medications are consistently associated with reduced risk of serious real-world outcomes, the magnitude of these associations have decreased over time alongside rising prescription rates. This underscores the importance of continuously evaluating medication use in different patient populations.” 

‍

Febrile seizure (FS) is a type of childhood seizure accompanied by a fever. It is not caused by infection in the central nervous system or other triggers of acute seizures. It is the most common form of childhood seizure, with an occurrence of 2% to 5% in all infants and children between 6 months and 5 years old. 

Noting that “To the best of our knowledge, no systematic synthesis of literature has assessed the nature and magnitude of the association between FS and ADHD,” a Korean research team performed a systematic search of the medical literature followed by meta-analysis to explore any such association. 

Meta-analysis of twelve studies with a combined total of more than 950,000 persons found that childhood febrile seizures were associated with 90% greater odds of subsequent ADHD. Correcting for publication bias reduced this slightly to 80% greater odds of subsequent ADHD. 

Limiting the meta-analysis to the subset of four studies with over 33,000 participants that adjusted for known confounders strengthened the association. Children who had febrile seizures had greater than 2.6-fold greater odds of subsequently developing ADHD. There was no sign of publication bias, but there was substantial divergence in individual study outcomes (heterogeneity). 

Further limiting the meta-analysis to two studies with a combined 654 participants in which clinical ADHD diagnoses were made by specialists – the gold standard – produced the exact same outcome. In this case, heterogeneity dropped to zero. 

The team concluded, “Overall, our systematic review and meta-analysis has shown a significant positive association between childhood FS and later occurrence of ADHD. Our findings add to the growing body of evidence questioning the notion that childhood FS are universally benign. In addition, the results highlight the need for longitudinal studies to better understand the association between FS and ADHD.”  

‍

ADHD treatment usually involves a combination of medication and behavioral therapy. However, medication can cause side effects, adherence problems, and resistance from patients or caregivers. 

Numerous systematic reviews and meta-analyses have evaluated the effects of non-pharmacological interventions on ADHD. With little research specifically examining game-based interventions for children and adolescents with ADHD or conducting meta-analyses to quantify their treatment effectiveness, a Korean study team performed a systematic search of the peer-reviewed medical literature to do just that.  

The Study: 

To be included, studies had to be randomized controlled trials (RCTs) that involved children and adolescents diagnosed with ADHD. The team excluded RCTs that included participants with psychiatric conditions other than ADHD.  

Eight studies met these standards. Four had a high risk of bias.  

Meta-analysis of four RCTs with a combined total of 481 participants reported no significant improvements in either working memory or inhibition from game-based digital interventions relative to controls. 

Likewise, meta-analysis of three RCTs encompassing 160 children and adolescents found no significant improvement in shifting tasks relative to controls. 

And meta-analysis of two RCTs combining 131 participants reported no significant gains in initiating, planning, organizing, and monitoring abilities, nor in emotional control. 

The only positive results were from two RCTs with only 90 total participants that indicated some improvement in visuospatial short-term memory and visuospatial working memory.  

There was no indication of effect size, because the team used mean differences instead of standardized mean differences.  

Conclusion:

The team concluded, “The meta-analysis revealed that game-based interventions significantly improved cognitive functions: (a) visuospatial short-term memory … and (b) visuospatial working memory … However, effects on behavioral aspects such as inhibition and monitoring … were not statistically significant, suggesting limited behavioral improvement following the interventions.” 

Simply put, the current evidence does not support the effectiveness of game-based interventions in improving behavioral symptoms of ADHD in children and adolescents. The only positive results were from two studies with a small combined sample size, which does not qualify as a genuine meta-analysis. All the other meta-analyses performed with larger sample sizes reported no benefits. 

‍