• A-Z
  • Directory
  • myUVM
  • Loading search...

Home – Child Mental Health Blog

Suicide Prevention in Schools

Posted: February 25th, 2015 by David Rettew

Suicide and suicide attempts remain a major public health problem.  There is now evidence that after years of decline, suicidal behavior is once again on the rise.  In Vermont, suicide is now the number two killer of older adolescents.  One can only imagine the attention and public health response that would occur if something like measles or ebola or terrorism was related to this kind of mortality here in Vermont and elsewhere. A recent compelling story on the subject was aired last week on WCAX by Darren Perron entitled Hidden Heartbreak Part 1 and Part 2.YAM

While there have been many initiatives taken to try and prevent suicidal behavior, it has remained a challenge to demonstrate that these efforts are effective, especially when it comes to decreasing discrete behaviors such as suicide attempts.  Since these are (thankfully) relatively rare events, very large sample sizes are needed to show the effect of an intervention.  A recent study, however, attempted to do just that by comparing the efficacy of three different programs in a sample that included over 11,000 adolescents from 168 different schools across 10 EU countries.  The paper was published in the prestigious journal, the Lancet.

The schools were randomized to administer one of three intervention programs.  The first, called Question, Persuade, and Refer focused on training teachers to identify and communicate with high risk youth. The second program, Youth Aware of Mental Health (YAM), was a universal intervention designed to teach all students to change negative perceptions and to enhance coping strategies through lectures, workshops, and educational materials.  The third program, Screening by Professionals, involved examining the study’s baseline mental health data and inviting those students who scored above established cutoffs to get a professional assessment.  There were also control schools that received educational posters only. The primary outcome measure was number of new suicide attempts and presence of severe suicidal ideation at 3 and 12-month follow-up, as assessed through self-report instruments.

At the three month follow-up, there were no significant differences between any of the intervention groups and the control condition.  However, at 12 month follow-up, students who received the Youth Aware of Mental Health (YAM) program had about half (0.70% versus 1.51%) as many suicide attempts compared to the control group and were half as likely to report serious suicidal ideation.  Putting the results another way, having 167 students in the program was found to be related to 1 less suicide attempt. These results did not vary according to sex or age, and there were no completed suicides that occurred during the study period.

The authors concluded that the Youth Aware of Mental Health program is effective in reducing the number of suicide attempts.  They urged more study and broader implementation of universal suicide prevention programs.

What makes this study exciting is that it is one of the first to document actual reductions in suicide attempt related to a school based intervention, as many studies prior to this have focused on increases in education and attitudes.  The YAM program does have a website for people interested in training and implementation of their method.  Doing a little research with help from the Vermont Youth Prevention group, it looks as though many Vermont schools utilize a program called Lifelines that has both primary and secondary prevention components.   This program, from a quick look, appears to have some overlap with the YAM program, although a more thorough review and comparison might be useful.


Wasserman D, et al. School-based suicide prevention programmes: the SEYLE cluster-randomised, controlled trial.  Lancet 2015: epub ahead of print.


Fewer Emergency Department Visits When Children Taking ADHD Medication

Posted: February 19th, 2015 by David Rettew

Children who meet criteria for ADHD are known to be prone to all types of injuries ranging from traffic accidents to colliding with objects at home.  While most of the studies on ADHD treatment concentrate on symptom reduction and school achievement, the small literature on accidents has been inconclusive.  A recent paper published in the journal Pediatrics, however, attempts to use a large database to examine this question more fully.

The authors utilized a large electronic database of health care information for patients living in Hong Kong.  The main outcome variable was the n

Copyright (C) 2010 Rajan Chawla / University of Vermont Medical Photography

Copyright (C) 2010 Rajan Chawla / University of Vermont Medical Photography

umber of trauma-related emergency department (ED) admissions between 2001 and 2013 for over 17,000 youth between the ages of 6 to 19 who at one time were prescribed methylphenidate.  The study design was such that each child served as his or her own control.  In other words,  ED admission rates were analyzed by comparing the number that occurred when the  child was taking medication versus time the same child was not taking medications. Incident rate ratios (IRR), similar to odds ratios, were calculated, controlling for age and season.  The authors also checked admission rates for ED visits not related to trauma as a control for their findings, under the hypothesis that methylphenidate would have no association in these cases.  They also examined their data in different ways to make sure their findings held.

Overall, they found that approximately 28% of children in the study had a trauma-related ED admission during the study period. The rate of admission during times children were being prescribed methylphenidate was significantly lower than during periods when the child was not prescribed methylphenidate, with an IRR of 0.91 (roughly a 9% reduction). This effect was found for both girls and boys and was particularly strong for older adolescents, where an impressive 32% reduction was found. Bolstering confidence for the results was that no association with methylphenidate was found for non-traumatic ED visits.  They also found similar results when they categorized their data slightly different ways.

The authors concluded that they were able to detect a protective association for methylphenidate related to traumatic injuries resulting in emergency department visits. The authors suggest that this factor be taken into account when deciding about treatment.

In looking at the raw numbers in their Table 1, it appears like these results would not have been significant had they not controlled for age and season. What that means isn’t exactly clear and the authors, unfortunately, did not comment on this. We also don’t know exactly what types of injuries did occur.

Lending additional confidence to the findings, however, relates to the study’s design in which each subject served as their own control.  The authors state that this aspect helped them reduce bias related to ADHD severity.  However, it is also possible that subjects more likely received medication when their symptoms were more severe.  If so, then this element would have biased the authors against finding the result that they did.  Finally, it should be noted that several of the authors did have financial ties to the pharmaceutical industry.

Lest anyone think that this post is all about advocating just for medications, I also can’t help but make reference to a similar study http://www.bmj.com/content/331/7531/1505 done in 2005 in England that showed an association between decreased ED visits for trauma and the release dates of two Harry Potter novels.  There is obviously more than one way to help children stay safe.


Man KKC, et al., Methylphenidate and the Risk of Trauma.  Pediatrics 2015:135:40-48.


Postconcussive Symptoms and Cortical Thickness in Hockey Players

Posted: February 9th, 2015 by David Rettew

There has been a lot of concern lately about concussions suffered all levels of sports competition. Some research exists that repetitive blows to the head can result in accelerated thinning of the cortex: a marker of possible compromised cytoarchitectonic integrity. This recent study published in the Journal of Pediatrics by the University of Vermont’s Dr. Jim Hudziak and coworkers examines the cortical thickness of ice hockey players as it related to a history of concussion and post-concussive symptoms.

Hockey - Photo by Steve Mitchell

photo by Steve Mitchell

The subjects for this study were 29 healthy male hockey players between the ages of 14 and 23, playing on preparatory school or collegiate ice hockey teams.   A total of 16 of them reported that they had been diagnosed with a concussion at least once by a medical professional. Five subjects in this sample had been previously diagnosed with ADHD. Current post-concussive symptoms were assessed using the Immediate Post-Concussion Assessment and Cogntive Testing (ImPACT) battery, a computerized assessment that probes a number of domains including memory, processing speed, impulse control, and reaction time.  As the name suggests, this instrument is often used soon after a concussive event. Thus, what makes this study more novel is that these players had not recently been concussed and were not currently complaining of symptoms.   Levels of behavioral problems were also assessed using the Youth Self Report and the Adult Self-Report instruments. Anatomical MRI scans were obtained with the main variable of interest being the thickness of the cortex in relation to the total score on the ImPACT, after controlling for age.

In terms of results, total concussion symptoms were significantly associated with a thinner cortex in multiple brain regions including the left dorsolateral, ventrolateral, and orbitofrontal cortices and the right dorsomedial cortex, and bilaterally in the tempoparietal cortices. These associations remained even after controlling for levels of presumably baseline behavioral problems. Additional analyses revealed that the brain changes were related especially to symptoms such as difficulty paying attention or remembering, fatigue, or the feeling of being in a mental fog.

Interestingly, a history of concussion by itself was not related to brain thinning overall, although a significant interaction was found between concussion history and age such that in several brain regions, subjects with no history of concussion exhibited age-related thinning in those areas while those with a history of concussion did not.   Another intruiging findings was that levels of attention problems, irregardless of cause,  were related to reduced thickness in the left anterior cingulate, left ventromedial prefrontal cortex, and left dorsomedial prefrontal cortex.

The authors report that this is the first study to find an association between brain structure and post-concussive symptoms in healthy male athletes.  They state that their findings related to brain changes and post-concussive symptoms in the absence of there being a direct link with concussion history suggests that it may be the more numerous and subtle blows that impact the brain more than discrete concussive episodes.

These data will likely add to the growing concern and effort to keep brains safe during sports so that the many benefits of exercise and team sport participation can be realized.


Albaugh M, et al.  Postconcussive Symptoms Are Associated with Cerebral Cortical Thickness in Healthy Collegiate and Preparatory School Ice Hockey Players. J Pediatrics, 2014, epub ahead of print.

Treating Parental Depression Benefits Children

Posted: January 28th, 2015 by David Rettew

At our clinic at the Vermont Center for Children, Youth, and Families, one unique component of our child psychiatry evaluations is the provision of also assessing mental health problems in the parents, using validated rating scales.  This element was included in the face of mounting data showing that successful treatment of psychiatric disorders in parents can result in behavioral improvements in their children.  Now, a recent study adds some new wrinkles to this issue while giving clues about how this effect might work.Parental depression

The study is a randomized double-blind 12 week clinical trial for 78 mothers with clinical depression that compares  two different antidepressants and their combination.  One group was treated with buproprion (Wellbutrin) only, one with escitalopram (Lexapro) only, and another group treated with their combination. Behavior problems in their 135 children, ages 7 to 17, were also assessed. Additionally, parenting behavior and maternal negative affectivity (guilt, hostility/irritability, anxiety) were also evaluated.

Treatment of the maternal depression was relatively effective with remission occurring in 67% of the sample.  There were no significant differences in remission between the treatment groups. However, the association between improvement in the mothers’ symptoms and improvement in their children’s depressive symptoms and functioning was significant for mothers in the escitalopram only group. In examining possible mechanisms for the child improvement, there was some evidence that mothers in the escitalopram group had improved abilities to listen and communicate with their kids. The children in this group corroborated the mothers’ report by noting greater maternal care and affection with treatment. Further, child scores improved among mothers with high levels of negative affectivity again for patients in the escitalopram only group but not for the other groups.

The authors concluded that the effect of children’s behavior improving as their mother’s depressive symptoms improve may depend of the type of treatment. They hypothesized that treatments that improve maternal anxiety and irritability may be especially helpful through their effect on parenting.

In discussing their results, the authors admit being somewhat surprised and puzzled with why the escitalopram only group seemed to fare better than the other two groups when it came to child behavior.  It also is important to note that this study does not imply that medications should be the sole focus of treatment for depressed parents, as another study found similar benefits when using interpersonal therapy (one of the evidence-based psychotherapies for depression).   Also worth noting is that this study was conducted before the FDA warning on ecitalopram that recommend not exceeding a dose of 20mg.  In this study, patients received as high as 40mg.

In summary, this study nicely helps connects the dots between changes in maternal symptoms, changes in child symptoms, and possible mechanisms for this effect through modifications in parenting behavior.  For primary care clinicians, it is important to ask about emotional-behavioral problems in parents when evaluating children.  Family physicians may particularly be in prime position to help their child patients by addressing the mental health concerns of the parents.



Weissman M, et al.  Treatment of Maternal Depression in a Medication Clinical Trial and Its Effect on Children.  Am J Psychiatry 2014, epublication ahead of print.

What Happened to Concerta?

Posted: January 9th, 2015 by David Rettew

There is quite a bit of confusion out there about the supply and availability of one of the staple ADHD medications, Concerta, which is the brand name of a long acting preparation of methylphenidate that uses a novel delivery system called OROS (Osmotic [Controlled] Release Oral [Delivery] System) to deliver the medication gradually throughout the day.

First of all, let me say that I have no affiliation and get no money from the pharmaceutical company, Janssen, that

OROS preparation contain the word "alza" on the capsule

OROS preparation contain the word “alza” on the capsule

makes the medication.  However, for me like many physicians out there, Concerta has become one of the first medications I use to treat ADHD, in conjunction with a variety of nonpharmacological interventions.

Concerta is now off-patent and thus generic preparations have been available for quite some time.  Some of these generics, in particular those made by Mallinckrodt and Kudco, didn’t use the same OROS technology and many patients have complained that these preparations don’t work as well as the brand name or another generic preparation by Watson/Actavis which makes an “authorized generic” preparation that does use the same OROS delivery (in fact it is exactly the same thing as Concerta but that’s another story).

In November, the FDA made a statement that their data suggest that the Mallinckrodt and Kudco generics are substantially different from Concerta to the point that they should no longer be considered therapeutically equivalent.   One specific concern is that these generics may release the drug too slowly over the course of the day.   As a result, the FDA no longer designated the Mallinckrodt and Kudco preparations as equivalent to Concerta (Mallinckrodt filed a lawsuit against the FDA about this but, again, that’s another story).   Practically, this means that if a physician writes for Concerta, the pharmacy no longer can just substitute this for one of these generics, as they have been doing for people with Medicaid and many commercial insurance programs that do generic switches when possible.

This gets to be a problem because a) the Watson/Actavis generic and even the brand name Concerta are in short supply and 2) the OROS preparations have dropped off the insurance company formularies and can’t be obtained without the patient and prescriber jumping though many hoops, including failing other generic long acting stimulants.  You can view the most recent version of Vermont Medicaid’s formulary here.

Here is what this all boils down to practically.

1. Prescribers need to write for “methylphenidate ER” if they are okay with patients getting the Mallinckrodt or Kudco preparations, which seems to be just fine for many people.  Remember, the problem was therapeutic equivalency not safety. An informal discussion among child psychiatrists in our shop revealed that problems with the generics were few.  However, you may want to watch out for the possibility that the medication won’t work well early in the day and then come on strong in the afternoon (the reverse patterns of what you usually see).

2.  If a prescriber writes for Concerta, then you could well get a call from the pharmacy because either they will have to dispense brand name Concerta if they can’t (or won’t) get the authorized generic (which generally seems to be the case).  That means the patient either pays the premium or the physician has to somehow get brand name Concerta approved by the insurance company.

3.  Prescribers need to switch to a completely different long acting stimulant that is on the formulary.

It is sad but the bottom line here that is that, at least for now, one of the main “go to” ADHD medications in its original form seems to be off the table for most families.  People are certainly complaining about this to the pharmaceutical companies, insurance carriers, and government agencies that are involved.  Hopefully this problem will be resolved soon.


Musical Training Linked to Enhanced Brain Maturation

Posted: December 2nd, 2014 by David Rettew

Patients who come to see child psychiatrists like Dr. Jim Hudziak at the Vermont Center for Children, Youth, and Families may leave with a prescription, but it often is not for a medication.  As part of a model he developed called The Vermont Family Based Approach (VFBA), there is increased emphasis on incorporating wellness and health promotion strategies into the overall treatment plan.  As Hudziak explains in a podcast related to the study, “One of my life goals is to see if there is a chance to move medicine away from its preoccupation with negative events and negative outcomes to argue that the opposite is also true, and that when positive things happen, positive outcomes will follow.”  Thus, the Violingoal of this model for children and families is to help them take steps not only to overcome whatever symptoms they have but to propel them towards true mental health and wellness.  To get there requires attention to domains such as nutrition, parental mental health, sleep, mindfulness, and physical activity, often given short shrift in traditional approaches.   Music and the arts are also highly encouraged within the VFBA.  According to the Department of Education, approximately 75% of American high school students rarely or never participate in music or art training outside of the school.

While participation in music and the arts is widely viewed as positive for child development, how it affects the brain remains only partially understood.  To investigate this question further and to bolster the scientific evidence behind the push for more involvement in music, Dr. Hudziak and his postdoctural associate Matt Albaugh, along with a team comprised of scientists from the University of Vermont, Montreal Neurological Institute, Harvard, and Washington University, examined brain scan data from the National Institutes of Health MRI Study of Normal Brain Behavior.   Their study was published as the lead article in the November edition of the Journal of the American Academy of Child & Adolescent Psychiatry.

The subjects for the study were 232 typically developing children without psychiatric illness between the ages of 6 and 18, all of whom received structural MRI scans at up to three different time points.   With these serial MRI scans the examiners were able to see how the thickness of the brain cortex changed with age.  Prior studies have indicated that the cortex generally thins across adolescence as the brain undergoes a normal “pruning” process that may be related to more efficient brain functioning.  A delay in this cortical thinning process, particularly in regions such as the prefrontal and orbitofrontal cortex, which are thought to be important for “executive control” functions such as inhibiting impulses and regulating attention, has recently been shown among those with clinical attention problems and ADHD.

The amount of musical training a child had was also measured to see if this variable interacted with age in its association to cortical thickness.  The average time playing an instrument was about two years.

Dr. Jim Hudziak

Dr. Jim Hudziak

The main result of the study was that years of musical training were indeed related to age-related cortical thinning. Specifically, more musical training was associated with accelerated thinning, not only in the expected motor cortices but also in some of the very same regions implicated in those with more pronounced attention problems. “What was surprising was to see regions that play key roles in emotional regulation also modified by the amount of musical training one did.”

The authors concluded that musical training was associated with more rapid cortical maturation across many brain areas, and they hypothesized that musical training may have beneficial effects on brain development for children whether or not they suffered from attention or executive function difficulties.

Certainly, much more research is needed to support the notion of musical training as an effective treatment for diagnoses such as ADHD, but this study raises some thought-provoking possibilities.   In the article, Hudziak and colleagues  highlight Venezuela’s El Sistema program that has brought musical training and performance to millions of disadvantaged children both abroad and here in the U.S.. Studies have shown important improvements in drop-out rates, employment, and community involvement among participants of the program.  Such efforts are critical as many families are unable to access music lessons due to their cost.   Dr. Hudziak, who has done research on the genetic influence of various traits and abilities, notes that our culture seems to have it backwards in promoting certain activities only for children who seem born to excel at them.  He questions why “only the great athletes compete, only the great musicians play, and only the great singers sing,” especially as children age.  He and his team have worked to improve local access to musical training through research studies and mentorship programs. The need is still high, however, and is now underscored by the increasing data linking wellness activities to measurable changes in brain development.


Hudziak JJ, Albaugh MD, et al.  Cortical thickness maturation and duration of music training: Health-promoting activities shape brain development.  JAACAP. 2014;11:1153-1161.

ADHD as a Brain Maturation Delay

Posted: November 19th, 2014 by David Rettew

Despite a wealth of evidence, the diagnosis of ADHD remains controversial, particularly outside of the medical community.  Some research has suggested that ADHD might be better conceptualized as a delay in brain maturation rather than a “disease state” per se.   However, more research is needed to support this hypothesis.

The brain imaging literature has increasingly moved away from studies that look at the size or activity of a particular region in isolation and towards the examination of  regional networks of several areas that are intrinsically connected to

ADHD related maturational lags.   Red and blue indicate areas of stronger connectivity lag

ADHD related maturational lags. Red and blue indicate areas of stronger connectivity lag

each other when the brain is doing tasks or at rest.  Perhaps the most studied of these networks relates to brain activity when it is not engaged in a particular task.  This has been known best as the default mode network or DMN.  The present study published in PNAS examines the development of these networks over time between individuals with and without a diagnosis of ADHD.

The study utilized functional MRI scans from a group of 135 individuals with a diagnosis of ADHD and compared them to 188 typically developing controls from ages 7 to 21. To investigate their hypotheses, the authors employed complex analyses called whole-brain connectomic methods which reportedly represent an advance from older techniques that used a small number of “seeds.”  These techniques allowed the authors to examine more than 400,000 different brain connections.

With these analyses, the authors did find maturation lags in those with ADHD, particularly within the DMN and  between the DMN and two other task positive networks (networks involved in specific cognitive functions), namely the frontoparietal and ventral attention network (VAN). The VAN is involved in salience processing, i.e. detecting the relevant stimuli the external environment while the frontoparietal network is involved in adaptive cognitive control.  The implications of both of these networks seem to make sense clinically when considering those with struggle with ADHD. 

The authors concluded that their data lend additional support that ADHD is related to important lags in brain maturation in areas that underlie the regulatory control of attention and behavior.

This is an important study published in a highly respected journal.  It is also quite methodologically complex and difficult to evaluate on a technical level for those who are not neuroimaging experts.  Nevertheless, the study provides ever more increasing and specific evidence that the challenges related to ADHD reflect “real” alterations in brain function and structure.

Indeed, the hypothesis of ADHD as brain maturation delay could offer a compromise between those inclined to dismiss the very existence of ADHD and those who view it from a more traditional disease model.  Before jumping on this train, however, I personally would like to see more evidence regarding whether the brain function of those diagnosed with ADHD eventually catches up or whether these differences persist late in life.  As far as I can tell, this study does not address this important point.


Sripada CS, et al., Lag in maturation of the brain’s intrinsic functional architecture in attention-deficit/hyperactivity disorder.  PNAS 111(39):14259-64, 2014.

Prevention Study Shows Promise in Reducing Psychopathology and Crime

Posted: November 12th, 2014 by David Rettew

One of the “Holy Grails” of psychiatry is the ability not only to treat existing behavioral problems effectively but to prevent them.  This task has proven challenging but remains the focus of many given the extraordinary human and financial costs associated with chronic psychopathology and criminal behavior.  Some people have become somewhat pessimistic about this goal, especially among at-risk children who at young ages already show signs of early conduct problems and rule-breaking behavior.

One project that has tried to intervene is called the Fast Track prevention program.  It began in 1991 as a multi-site effort to test whether comprehensive early intervention could prevent later psychopathology and criminal behavior in a

Rates of Clinical Problems at Age 25 for Intervention and Control Participants

Rates of Clinical Problems at Age 25 for Intervention and Control Participants

group identified as at-risk. A total of 979 kindergarteners from 4 geographic areas were identified as showing early conduct problems, based on several rating scales. They were then randomized into a control group and an intervention group designed to increase social competence through skills training, parent behavior management training, peer coaching, and academic support over a period of 10 years. When the subjects were 25 year old, they were reassessed for their arrest records as well as for psychopathology, using standardized rating scales.

As young adults, the rate of a psychiatric or substance abuse diagnosis was 69% in the control group compared to 59% in the intervention group which was a significant difference. This result held for different demographic and severity groups. Significant group differences were also found related to crime, risky sexual behavior, and self-esteem.  One key metric was related to arrest and conviction.  The incarceration rate was 6.3% for the control group versus 5.0% for the intervention group.  While this difference was not statistically significant, there was a significant drop of severity-weighted violent and substance-related crime conviction of 31% and 35%, respectively. Significant effects were not found related to graduation rates or employment.  Disappointingly, there was also little effect on how these subjects parented their own offspring.

The study authors concluded that their intervention did result in a significant reduction in psychopathology and criminal behavior.  They advocate strongly that prevention efforts can succeed and should be encouraged.

Of note, this program was reported to cost approximately $58,000 per student and future cost analyses are planned.  In the discussion, they compare that amount to the estimate that the cost of chronic criminality is  5.3 million dollars per person.  The authors point out that one limitation of the study was that they were unable to determine which type of intervention was most protective among the many different types that were tried.

In looking at these data, I have to say that I was both inspired to see a program show clear results but also humbled by what I think many people will find as rather modest results from a fairly intensive and long intervention.   It takes an awful  lot of work to drop the rate of psychopathology from about 70% to 60% or to reduce the rate of certain types of crime by 30%.  This effort seems well worth doing in my view, despite an appreciation for how powerful the forces can be that conspire to keep at-risk children down.


Conduct Problems Prevention Research Group.  Impact of Early Intervention on Psychopathology, Crime, and Well-Being at Age 25.  Am J Psychiatry, 2014, epub ahead of print.

Omega-3s Founds to Improve Child Aggression

Posted: November 3rd, 2014 by David Rettew

 The benefits of Omega-3 supplementation has been touted for a wide range of therapeutic and health promotion uses.  While there is emerging data for problems such as ADHD, the literature has still suffered from issues such as small sample sizes, lack of randomization, short duration, and lingering questions about optimal dose.  This recent study sought to address some of these limitations using a randomized double-blind placebo controlled design and studying children for a total of 12 months, which included 6 months of study after the supplementation ended.  Another innovation for this study was the additional measurement of parent behavioral problems, under the notion that these could Omega-3mediate improvement in child behavior.

The nonclinical sample included 200 children from the ages of 8 and 16 from the island nation of Mauritius.  For those of you without a PhD in geography, this is a small island in the Indian Ocean off of Madagascar (yes I had to look it up too).  Half of the sample was radomized to receive 1 gram of Omega-3s (300 mg of DHA, 200 mg of EPA, 400 mg of alpha-linolenic acid, and 100 mg of DPA) delivered in a fruit drink while another 100 received a fruit drink without the Omega-3s. Behavior problems were measured by parent- and child-report at baseline, at the end of the six month study, and at 12 months, using our favorite instruments the Child Behavior Checklist (CBCL) and Youth Self-Report, supplemented with other measures of aggression.  As mentioned, an interesting aspect of this study was that rating scales were also given to parents to examine their own levels of psychiatric symptoms both at baseline and at follow-up.

The main finding was a significant effect for omega-3 supplementation across a wide range of parent-reported child behavior.  Improvements were found not only in the predicted areas of aggression and externalizing problems but also for internalizing problems such as anxiety and depressed mood.  The changes for child self-report behavior were less dramatic but present for things like both reaactive and proactive aggression.  Indeed, even the troubling and hard to treat callous-unemotional traits showed improvement by parent-report.  For many measures, significant differences were mainly apparent at the 12-month interval, six months after the trial ended, thus emphasizing the need to stay with treatment over a long period of time.  Overall, externalizing behavior decreased 41.6% six months after the trial ended compared to a drop of around 11% for placebo. The overall effect size was found to be moderate (d=-.59).

Also extremely interesting and providing further evidence for a family-based approach to child mental health is the finding that parents also showed reductions in measures of their own psychiatric symptoms (even though they weren’t taking the supplements).  Furthermore, improvement in parental symptoms was found to substantially mediate the improvement found in the child’s behavior.  An impressive 60.9% of the improvement in child antisocial behavior, for example, could be attributed to reductions in the parents’ reduction in psychopathology.

The authors concluded that their data provide support for the utility of using omega-3s to reduce both internalizing and externalizing behavior and suggest that one mechanism through which children get better is that their parents improve with regard to their own psychopathology. 

In my mind, this is an important study in many ways and I’m surprised it wasn’t covered more widely.  This may have been because it was published in a certainly reputable but not very prominent journal.  What is remarkable about this study is not only the fairly robust improvement noted with Omega-3 supplementation but also the demonstration of how important it can be to improve parental symptoms in the pathway of improving child behavior.

At the same time, some limitations are worth noting.  The sample was non-clinical and the authors did not examine whether or not more symptomatic children responded to the Omega-3s the same way that less symptomatic children did. Also, there obviously will be some questions about how generalizable this sample is coming from a fairly remote island.   Finally, it needs to be said that the commercial company that provided the Omega-3 drinks, a Norwegian company called Smartfish, supported this study financially, and it is important that we give that fact the same skeptical eye that we would apply if we were talking about a prescription medication study supported by a pharmaceutical company.

Nevertheless, these results are important and add to the growing body of research suggesting that Omega-3s should be on our radar screen as clinicians.  The specific dose is also helpful as a guide.


Raine A, et al. Reduction in behavior problems with omega-3 supplementation in children aged 8–16 years: a randomized, double-blind, placebo-controlled, stratified, parallel-group trial.  J Child Psychol Psychiatry. 2014, epub ahead of print.

Type 2 Diabetes Risk with Antipsychotic Medication Treatment Reported at 0.72%

Posted: October 15th, 2014 by David Rettew

While the  metabolic risks associated with antipsychotic medications use are now well known, finding some actual numbers about the risk have been slow to come.  A recent study by Nielsen and colleagues published in the Journal of the American Academy of Child and Adolescent Psychiatry offers some specific numbers that can be used when discussing the potential risk of developing Type 2 Diabetes in the course of treatment with antipsychotic medication.JAACAP logo

To arrive at these numbers, a case-control design was used that obtained information from a large Danish registry of psychiatric patients under 18 years of age seen from 1999 to 2010.  The indicator of having Type 2 Diabetes was the prescribing on an oral antidiabetic medication comparing between those who were and were not also prescribed an antipsychotic medication.  Antipsychotic use was counted as positive if one or more prescriptions were filled. Regression analyses were used, controlling for age, sex, and diagnosis in order to assess for the possibility that the diagnosis, rather than the treatment, might be behind any increase in diabetes. The records of nearly 50,000 psychiatrically ill youth were examined.

Overall, the rate of Type 2 Diabetes among youth treated with antipsychotics was 0.72% compared to a rate of 0.27% in psychiatrically ill youth not given antipsychotic medication. This difference, when controlling for potential confounds, resulted in a significantly elevated odds ratio of 1.60.  Being female and being older at the time of diagnosis was also related to diabetes. When metformin was excluded as an indicator of diabetes (because of the possibility that it was being used in many cases to treat wait gain and protect against the development of diabetes), the odds ratio related to antipsychotic use went up to 3.71; however, this odds ratio was no longer significant due to such a small number of cases.  (The authors state in the discussion that metformin is rarely used in Denmark for weight control).  Overall, the number needed to harm was calculated at 224.

The authors concluded that antipsychotic medication use does increase the risk of developing Type 2 Diabetes.  In the Discussion section, they recommend close adherence to indication and metabolic monitoring guidelines.

It is nice to have some specific numbers to talk about when having discussion of potential side effects that don’t rely on small and short-term clinical trials.  At the same time, however, these kinds of large registry databases can struggle with measuring other factors that may be important, such as compliance with treatment and overall duration. If milder diabetes was diagnosed but controlled through nonpharmacological means, for example, this study would have misclassified these children.

It will also be interesting to see how this Diabetes risk will be interpreted.  An 60% increase sounds like a lot, yet the overall rate was 0.72% which is likely lower than many people were expecting.  Several of the authors have rather extensive ties to various pharmaceutical companies, which will likely cause some people to discount the results.  As far as I can see, the study has not received much press at all, perhaps because the results were not that dramatic in either direction.


Nielsen RE, et al.  Risk of Diabetes in Children and Adolescents Exposed to Antipsychotics: A Nationwide 12-Year Case-Control Study. JAACAP. 53(9):971–979, 2014.


Contact Us ©2010 The University of Vermont – Burlington, VT 05405 – (802) 656-3131
Skip to toolbar