VIENNA – The creation of a simple risk score that accurately predicts which adolescents in the general population will develop persistent substance dependence as adults has been one of the highlights of the year in addiction medicine, Wim van den Brink, MD, said at the annual congress of the European College of Neuropsychopharmacology.
“These predictors are not very difficult to assess. Clinicians will be interested to know that the positive predictive value of the screen is threefold greater than the persistent prevalence rate,” noted Dr. van den Brink, professor of psychiatry and addiction at the University of Amsterdam and director of the Amsterdam Institute for Addiction Research.
Bruce Jancin/Frontline Medical News
The New Zealand researchers developed what they call “a universal screening tool” by working backward in an analysis of a representative group of 1,037 individuals born in Dunedin, New Zealand, in 1972-1973 and prospectively followed to age 38 years, with a 95% study retention rate. Along the way, participants were assessed for dependence on alcohol, tobacco, cannabis, or hard drugs at ages 21, 26, 32, and 38.
Persistent substance dependence in adulthood, defined as dependence at a minimum of three of the assessments, was present in 19% of subjects.
The investigators found that the presence in childhood or adolescence of any four of nine risk factors had an area under the curve of 80% for persistent substance dependence as an adult. The sensitivity was 43%, with a 93% specificity. The positive predictive value was 60%, and the negative predictive value was 87% (Psychol Med. 2016 Mar;46[4]:877-89).
The nine risk factors are low family socioeconomic status, a family history of substance dependence, childhood depression, childhood conduct disorder, early exposure to substances, adolescent frequent alcohol use, adolescent frequent cannabis use, male gender, and adolescent frequent tobacco use.
The single least potent predictor was low family socioeconomic status, with an associated 1.73-fold increased risk. The strongest predictors were adolescent frequent tobacco use, which conferred a 5.41-fold increased risk; adolescent frequent cannabis use, with a 4.25-fold risk; and childhood conduct disorder, with a 3.2-fold increased risk.
The investigators also analyzed the screening tool’s performance in predicting a modified outcome consisting of adult persistent dependence on any of the target substances except for tobacco. The predictive power of having any four of the risk factors was similar to that found in the main analysis; however, the two strongest predictors now became adolescent frequent cannabis use, with a 9.5-fold increased risk, and childhood conduct disorder, with a relative risk of 5.42.
Regarding childhood conduct disorder as a risk factor, Dr. van den Brink said, “If you are a child with conduct disorder, your chances of becoming substance dependent in coming years is more than fivefold greater than in a child without conduct disorder.”
This raises the question of whether effective treatment of childhood conduct disorder might prevent later development of persistent substance dependence in adulthood. The answer remains unknown. Although there is no approved drug therapy for conduct disorder, methylphenidate is widely prescribed, especially in young patients with comorbid attention-deficit/hyperactivity disorder.
Several years ago a meta-analysis of 15 longitudinal studies with more than 2,500 participants concluded that stimulant therapy of childhood ADHD neither increased nor reduced the risk of subsequent substance use disorders (JAMA Psychiatry. 2013 Jul;70[7]:740-9). Prescribing physicians were happy to hear they weren’t causing iatrogenic injury, but Dr. van den Brink said he was never comfortable with the investigators’ conclusion.
“There was a lot of heterogeneity in the data, so the overall conclusion might not be the best conclusion,” he said.
He said has become more convinced of that than ever as a result of a recent randomized, double-blind, placebo-controlled MRI study of cerebral blood flow in response to methylphenidate in stimulant-naive patients with childhood or adult AHDH. The investigators found that MRIs obtained 1 week after the conclusion of 16 weeks of methylphenidate therapy showed increased blood flow in the strial and thalamic areas in the pediatric ADHD patients but not in the adults with ADHD (JAMA Psychiatry. 2016 Sep 1;73[9]:955-62).
This is evidence of an age-dependent sustained effect of methylphenidate therapy on dopamine striatal-thalamic circuitry in children that’s not related to the drug’s clinical effects, which were gone after a week off therapy. The question is, Does this effect represent neurotoxicity, or is it an expression of enhanced brain maturation? Dr. van den Brink said he suspects it’s the latter but cannot exclude the former possibility.