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

Home – Child Mental Health Blog

Childhood Bullying Linked to Adult Chronic Inflammation

Posted: July 14th, 2014 by David Rettew

Can getting bullied make you sick?  Bullying is getting a lot of attention and perhaps deservedly so.  Research continues to show that being bullied in childhood can have profound and long-lasting negative effects both behaviorally and relative to other health outcomes. A recent study by Bill Copeland (a psychologist who, by the way, received his PhD at UVM and was a graduate student of Jim Hudziak back in the Bully 2day) and others from Duke generated some headlines by exploring a potential mechanism of these associations, namely chronic inflammation, that may connect the dots between being a bully victim and poorer health as an adult.  The study was published in the journal Proceedings from the National Academy of Sciences (PNAS).

The study uses data from the Great Smoky Mountain Study (GSMS). While many primary care clinicians may have never heard of this study, the GSMS has tracked the lives of a group of children living in rural North Carolina for many years, and has been a major source of epidemiological data.

Based on child and parent report, children were divided up into the typical categories of not being associated with bullying (the biggest group), bullies, victim of bullies, and those who were both bullies and bully-victims.  These subjects are now young adults and levels of  C-reactive protein (CRP),  a known marker for how much chronic low grade inflammation a body experiences, were measured.

Across childhood and adolescence, CRP tends to rise for everyone, However, those who were bullied more often were found to have higher than expected CRP levels compared to those uninvolved in bullying. Perhaps even more unexpected, however, was that those who tended to bully others had lower CRP levels. Finally, individuals who went back and forth between being a bully and being a victim had CRP levels that were no different than those not involved with bullying.

Copeland and his coauthors concluded that chronic inflammation may be a key factor involved in the poor mental and physical health that has been well documented among victims of bullying. They hypothesized that the increased inflammation might be due to bully victims experiencing less of the anti-inflammatory effects of cortisol which can become less responsive under conditions of chronic stress.

One thing that did not fit so well in this study was that their group of bully-victims that had the same CRP levels of those uninvolved in bullying. This group, according to other published studies, often has been shown to have the worst mental and physical health of all, so one might expect that they would have had high inflammation as well.

Aside from demonstrating how the worlds of physical and mental health are so interconnected, this study provides some important insights into a possible avenue through which bullying may exert its negative effects.

To learn more about bullying you can look at the government website stopbullying.gov.


Copeland W.  Childhood bullying involvement predicts low-grade systemic inflammation into adulthood. PNAS 2014;111(21):7570–7575.

Is Universal Suicide Screening in Primary Care Worthwhile? Nobody knows

Posted: July 1st, 2014 by David Rettew

Suicide remains a leading cause of death and is a major public health concern. Studies have demonstrated that many individuals who die by suicide often see their primary care physician soon before the event.  Thus, it is important to know whether suicide screening might potentially prevent some of these tragic deaths.  To that end, the U.S. Preventive Services Task Force,  an organization that issues guidelines regarding prevention measure in medicine, looked at the evidence once again ten years after an earlier report when the said that they could not make a recommendation for or against the practice due to a lack of data.USPTF

Their analysis attempted to focus on adolescents, adults and older adults who were not at an elevated risk for suicide at baseline and did not have an identified psychiatric disorder. The task force attempted to find studies that addressed three areas, namely 1) the accuracy of suicide screening tests, 2) the effectiveness of interventions to decrease suicide, and 3) potential negative effects of suicide screening and treatment.

When it came to results, the authors found only four screening studies on suicide screening, all of which used a different instrument.   Not enough data were available to determine if the screening was worthwhile, although the report didn’t really summarize these studies. The task force also found a general lack of evidence regarding the efficacy of treatment and any potential negative effects of suicide screening or treatment.

The overall conclusion of the task force was that, once again, the current database is insufficient to be able weigh the relative benefits and risk of preventive suicide screening in primary care. The task force recommended additional research to fill these significant gaps.

One might wonder in reading this often hard to follow report is why they chose to issue it again if the final conclusion of “insufficient data” remains unchanged.  Perhaps it was to spur additional interest and attention.  It is important to note that these recommendations (or lack thereof) do not suggest that there are no significant risk factors for suicide worth identifying. Indeed, the group has previously issued a recommendation that screening for depression, probably the strongest risk factor for suicide, be performed in primary care offices.


Lefevre M.  Screening for Suicide Risk in Adolescents, Adults, and Older Adults in Primary Care: U.S. Preventive Services Task Force Recommendation.  Annals of Internal Medicine 2014;160(10):719-727.

DCF and UVM Create New Program to Improve Antipsychotic Medication Prescribing

Posted: June 18th, 2014 by David Rettew


A new combined effort between the Vermont Department of Children and Families (DCF), the Vermont Department of Mental Health, and the Child & Adolescent Psychiatry Fellowship Program at the Vermont Center for Children, Youth, and Families (VCCYF) will support caseworkers who often need to make complex medical decisions regarding the children under their care.

While it has been well known that the number of youth who take antipsychotic medication has been increasing nationwide, the rate tends to be about five times higher among youth in state custody.  Antipsychotic medications can be an important component of treatment for some, yet these medications also carry the potential for serious risks and side effects including movement disorders, obesity, diabetes, and high cholesterol.  When kids enter state custody, DCF caseworkers become responsible for consenting to these and any other medications and medical decisions.  This responsibility can be a challenge due to the large caseloads of the social workers and the fact that these children often move around to various placements around the state.  Unlike most parents, DCF caseworkers are often unable to attend physician appointments with their children which is where the risks and benefits of medications are typically discussed.  DCF caseworkers are also not clinically trained in pharmacology.

To support DCF caseworkers in this challenging task, a new program will begin next month that will help them in deciding whether or not a antipsychotic medication that is being considered for a child in custody is appropriate.  This includes the following:

1.  Additional training provided to the caseworkers about antipsychotic medications and their potential risks and benefits.

2.  A more rigorous written informed consent process that will be required between the DCF caseworker and prescribing clinician to ensure that children in DCF custody who are being prescribed antipsychotics are getting these medications for the right reasons and being monitored according to best practice guidelines.

3. The opportunity for caseworkers who may have questions or concerns about the consideration of antipsychotic medications for one of their clients to get an independent opinion from one of the child psychiatry fellows in the VCCYF training program who is supervised by UVM child psychiatry faculty.   These free consultations can occur whenever the caseworkers would like one and are required in certain situations (such as a child under six years old).

The project has been led at DCF by Cindy Walcott, Deputy Commissioner for Family Services.   We are excited to be part of this effort both to help make sure these vulnerable children are getting the best care possible and to give our child psychiatry fellows a valuable educational experience.  Vermont has already seen a decrease in the use of antipsychotic prescriptions among youth over the past several years.   Improving access to evidence-based psychotherapy for oppositional and aggressive children, educating prescribers on when and how to reduce or discontinue these medications when appropriate, enhancing the flow of medical information between different practices, and using technology to remind physicians about the need for regular labwork can all help the prescribing of this class of medications to be closer to best practice recommendations.

The new policy begins on July 1.  Primary care clinicians who prescribe antipsychotic medications to children in DCF custody can expect to be asked to complete this informed consent process with the child’s caseworker over the next few months.


Behavior Problems and Screen Time: What Causes What?

Posted: June 2nd, 2014 by David Rettew

There have been many studies that have demonstrated links between excessive childhood screen time and negative outcomes including attention problems and aggression. Most of them imply causation but generally can’t prove it because the studies don’t measure child behavior before the screen time occurs.  Yes playing Minecraft for 6 hours per day might cause problems, but it could also be true that kids whose have attention problems are drawn more to the bells and whistles of video games. A recent study published in the journal Pediatrics examined this question in relation to poor infant self-regulation and media usage at age 2.Ipad in crib

The study comes from the Early Childhood Longitudinal Study – Birth Cohort and included 7450 children from the community, oversampled for higher rates of ethnic minorities. Infants were assessed for the regulatory abilities at an age of 9 months using a shortened version of the Infant Toddlers Symptom Checklist.  Items involving fussiness, needing constant attention, and sleep problems loaded onto the overall score of regulation problems.  Media usage was measured at age 2 by simply asking caregivers about average media use.  Regression analyses were performed to examine the link between infant regulatory abilities and media use, controlling for other potential confounds.  The authors examined the data both quantitatively (i.e. using the actual numbers of media time) and categorically dividing toddlers into groups that watched more or less than 2 hours per day (based on the AAP guidelines of two hours per day maximum of screen time).

Overall, a small but statistically significant link was found between infant regulatory abilities and age-two media use. The analyses showed that children with poorer regulatory skills (which was about 39% of the sample) consumed approximately 14 more minutes (9 minutes when adjusted for other variables) of media per day than those with good regulatory skills. At age 2, children averaged about 2.3 hours of media use per day with 40% of the sample going above the recommended 2 hours per day. The association between dysregulation and media use was stronger for children who remained dysregulated at age 2 and for children from English speaking and lower SES families.

The authors’ conclusions were somewhat unexpected.  Despite showing evidence that dsyregulation might be a cause rather than a result of media use, the authors argued that the amount of use associated with infant dysregulation was too small to account for the finding from other studies that excessive screen time is linked to negative child behavior.   They cautioned about a cycle in which fussier infants are more likely to be placed in front of screens (to give parents a break or occupy them in a safe place) which then in turn exposes them to the negative effects of increased screen time and thus reinforces the initial problem.    While this is certainly a plausible hypothesis, it requires a study that assesses child behavior, parent behavior, and screen time  at multiple time points.   We are still waiting for that one, as far as I know.


Radesky JS, et al.  Infant Self-Regulation and Early Childhood Media Exposure.  Pediatric; online publication April 14, 2014



CDC Launches New Website on Parenting of Young Children

Posted: May 27th, 2014 by David Rettew

by Allison Hall, MD

Allison HallPrincipals of good parenting like paying attention to positive behavior and using time out may sound easy but can be difficult to put into practice.   The Center for Disease Control (CDC) has recently created a website on parenting young children with some excellent explanations, useful forms such as schedules and charts as well as videos which illustrate parenting techniques which are supported by decades of research. The videos, which have a humorous touch, illustrate pitfalls we often tumble into as parents and show concrete examples of how to communicate positively, give clear instructions, create structure, and set rules and use consequences including time out.

The site has a very supportive, non-critical tone.   It recognizes that all children are different and there is no one size fits all method of parenting. The short videos could be an excellent jumping off place for discussing the use of these techniques with parents.

CDC Parenting



Patches of Brain Disorganization Identified in Autism

Posted: May 20th, 2014 by David Rettew

Characterizing the precise pathology in autistic spectrum disorders, let alone other psychiatric disorders, has remained challenging. Postmortem studies of actual brain tissue are rare but have the potential to provide important clues regarding specific abnormalities in neurodevelopment that are not detectable through other means such as neuroimaging. This major study published in the New England Journal of Medicine provides direct and provocative evidence of specific neuronal pathology that may underlie some of the core features of autism. Patch

The authors used RNA in situ hybridization techniques in order to look for gene expression patterns of 25 potential genes that have been implicated in the pathogenesis of autism and mark the presence of particular types of cells at different layers in the cortex. These analyses were performed on areas within the frontal, temporal, and occipital lobes of 11 children with autism and 11 children controls from two established brain banks. Children were between 2 and 15 years of age and the autistic sample was predominantly male.

In 10 out of 11 children with autism, compared to 1 out of 11 children without it, the study found focal areas of significant cytoarchitectural disorganization in the prefrontal and temporal cortex but not in the occipital cortex. These areas were referred to as “patches” and tended to be between 5 and 7mm long, often existing right next to unaffected cortex. The dysfunction was found for different types of neurons but not glial cells. While there was variability found in terms of the precise types of cells and cortical layers that were most affected in the autistic group, such a relatively unique and consistent finding is a major step forward towards understand the precise brain differences in children with autistic spectrum disorder.   The data also strongly points to autism beginning in the prenatal period  when cortical layering and cellular differentiation are occurring.

The authors wrote that they were somewhat surprised to see such as similar pathological feature across so many children with autism given its heterogeneous nature. They also note that while the presence of these patches was evident, their study does not suggest a mechanism for how these abnormalities in neuronal migration and differentiation they develop.

For those interested in visualizing these brain changes, the NIMH released a YouTube video that shows what the patches look like.  I would definitely recommend looking at this to better flavor of these intriguing findings.


Stoner R, et al. Patches of disorganization in the neocortex of children with autism. NEJM 2014;370:1209-1219.


Vermont Youth Prescribed Fewer Antipsychotic Medications

Posted: April 28th, 2014 by David Rettew

Newly analyzed data from the Department of Vermont Health Access is revealing important trends regarding psychiatric medication use in Vermont children and adolescents.  The data examines Medicaid insured youth who took at least one medication in various classes (antidepressants, ADHD medications, antipsychotics, etc.) over the course of the year, comparing rates between 2009, 2011, and 2013.  Efforts were made to ensure that children were counted only if the medication was used for psychiatric conditions rather than other causes such as seizures or allergies.   Separate statistics were computed for children within the foster care system.Antipsychotic Meds 2009-2013

One of the highlights is a decrease in the use of antipsychotics among youth.  Among 6-12 year olds, the rate dropped from 3.1% in 2011 to 1.7% in 2013, a 45% drop.  For adolescents, the rate went from 4.8% in 2009 to 3.5% in 2013.  Among foster care children aged 6-12, the rate drop was even more dramatic (17.2% in 2009 to 7.2% in 2013).   Antidepressant rates were relatively flat, while ADHD medications increased somewhat.  For example, the percentage of adolescents aged 13-17 taking ADHD medications such as stimulants increased from 12.10% in 2009 to 14.1% in 2013, while for younger children (6-12 years old) the rate was pretty stable (13.0% in 2009 to 13.4% in 2013).

These data continue to support the conclusion that the pattern of psychiatric medication usage in Vermont youth is different than most parts of the country.  While it is unclear what is responsible for these trends, many efforts have been launched across the state by individuals here at the Vermont Center for Children, Youth, and Families and elsewhere to try an ensure that medications are used carefully and only as a component of a more comprehensive treatment plan that includes many health promotion strategies.

Treatment Gains Persist in Childhood Anxiety Disorders

Posted: April 11th, 2014 by David Rettew

The Child/Adolescent Anxiety Multimodal Study (CAMS) represents one of the largest and most comprehensive studies to date on the comparative efficacy of different types of treatment for anxiety disorders. In the original NIMH funded study, a total of 488 youth with various combinations of separation, generalized, or social anxiety disorder were randomized to treatment groups consisting of 12 weeks of cognitive-behavioral therapy, medication with sertraline, combination medication and CBT, or pill placebo. At the end of the study, combined treatment was found to be superior to either active treatments alone and all treatments were superior to placebo.CAMS

A new study now describes the participants’ level of functioning and symptoms approximately 6 months later.  In the interval from study completion to follow-up, those who responded at the end of the original study were given 6 monthly booster sessions in their assigned modality. They were, however, permitted to add other types of treatment. Those that had originally received placebo were offered treatment if they did not respond to placebo, but were not included in these follow-up analyses.

One piece of good news overall was that, across all treatment groups, about 80% of youth who had originally responded to treatment kept their response at follow-up.  In terms of group differences, those in the combined group continued to do somewhat better than those in the CBT alone or sertraline alone groups on quantitative assessments of symptoms, although with categorical designations of responders versus nonresponders, the rates did not differ statistically.  This was due to the CBT and sertraline groups increasing their response rate (to around 70%) while the combined group kept its response level roughly the same (around 80%). The CBT alone and sertraline alone groups continued not to differ from each other.

The authors concluded that successful treatment of anxiety disorders results in a sustained response over the next 6 months for most individuals. They speculated that the lack of separation between groups over time may be related to the naturalistic design and the greater use of combination medication and therapy treatment among many subjects.

These results offer solid evidence that the vast majority of youth with anxiety disorders can improve significantly with different treatments and can maintain those gains at least over the next 6 months.  Their finding that the separations between groups erodes somewhat over time mirrors similar studies of other disorders where patients are followed naturalistically after the randomization stops. Overall, there is evidence that combination treatment can be the fastest path to improvement, although single modality treatment using evidence based practices may catch up.  Current treatment recommendations continue to be that childhood anxiety disorders should be treated with cognitive-behavioral therapy on its own or with medications for more severe cases.

Piacentini J, et al.  24- and 36- week outcomes for the child/adolescent anxiety multimodal study (CAMS).  JAACAP 2014;53(3)297-310

Study Finds Link Between In-Utero Acetaminophen Use and Child ADHD

Posted: April 1st, 2014 by David Rettew

A recent study in JAMA Pediatrics caused quite a bit of a stir in finding a possible link between acetaminophen taken during pregnancy and later childhood ADHD.   Acetaminophen has been one of the few medications that has been thought to be pretty safe to use during pregnancy and thus has been the “go to” agent for pain, fever, and other indications, especially given the known risks of some of the alternatives such as NSAIDS.  Now, however, some people aren’t so sure.Tylenol

The study used data from the Danish National Birth Cohort and assessed over 64,000 pregnant mothers.  The amount of acetaminophen use was measured from computer-assisted telephone interviews.  Child behavioral data was collected at age 7 using scores from the Strengths and Difficulties Questionnaire as well as medical records indicating a child took stimulant medications or had an ADHD diagnosis (in this study it was actually ICD hyperkinetic disorder but it is very similar).  Hazard ratios, similar to odds ratios, were calculated to compare the risk of ADHD between acetaminophen exposed and non-exposed mothers, controlling for some potential confounding factors such as maternal infections, inflammation, and maternal mental health.

The study found that majority of mothers did use acetaminophen at some time point during pregnancy.  Compared to mothers who did not use acetaminophen at all, the risk of ADHD in the offspring of mothers who did was significantly elevated, with hazard ratios ranging between 1.13 and 1.37 depending on how ADHD was measured. As an example, elevated scores related to hyperactivity were found in 5.7% of children whose mothers took any acetaminophen during pregnancy versus 4.3% of children who mothers did not. There was also some evidence that higher overall acetaminophen use and use during later pregnancy was associated with higher ADHD risk.

The authors concluded that in utero acetaminophen exposure was associated with increased rates of ADHD or ADHD-like behaviors and they urged further study to verify the findings.  The journal also featured an accompanying editorial which  recommended that  clinical practice not be fundamentally changed from this single non-controlled study.

The media, for the most part, also seemed to get that message and generally did a decent job of not going overboard with their headlines or text.  This was probably good given that everyone seemed to cover the study about a month ago (even I got asked to comment for ABC news).  Clearly, the biggest potential concern is the very real possibility that it is the factors that cause a pregnant mother to take acetaminophen, and not the drug itself, that underlies the link with ADHD.  While the authors took some steps to account for this possibility, they acknowledged that they could not completely control for it.  As there is also some preliminary data linking things like inflammation and infections during pregnancy and child behavioral problems, some restraint prior to changing recommendations seems appropriate.   It is one thing to try something that at best is harmless to see if a health benefit can be obtained (even in the absence of great data), but it is quite another scenario when those actions (or lack of action) may make things worse.   We will have to see, with the message now perhaps being that no medications, over the counter or not, should be taken casually while pregnant.

What might be the mechanism behind this association?  Nobody knows, but the authors did speculate about the possibility that acetaminophen might alter sex or thyroid hormone levels.



Liew Z, et al.  Acetaminophen Use During Pregnancy, Behavioral Problems, and Hyperkinetic Disorders.  JAMA Pediatrics. 2014:online publication.

ADHD Diagnosis Rising Nationally, but not in Vermont

Posted: March 5th, 2014 by David Rettew

The diagnosis of ADHD  has been getting a lot of media attention lately over concerns about possible overdiagnosis and overtreatment.  A recent study from the CDC’s National Survey of Children’s Health now provides important national statistics regarding the prevalence of ADHD, as reported by parents, and the trends over time.  Reports from this survey previously found an increase of 22% from 2003 to 2007 with an overall rate of 9.5%.  At that time, 4.8% of U.S children and adolescents were taking an ADHD medication.  This new study provides information up to 2011.ADHD prevalence by state

The data come from a random-digit-dialed telephone survey of parents that was performed in over 75,000 households using both land lines and mobile phones.  Parents were asked about lifetime and current diagnosis of one of their children as well as about severity, age of onset, and treatment.  The overall response rate was 23.0%.  Analyses were weighted to estimate a total prevalence.

Overall, the rate of ever having ADHD among 4 to 17 year olds nationally was 11% (15.1% for boys, 6.7% for girls), representing 6.4 million children. This rate represents an increase of 42% from 2003 to 2011. Of those will current ADHD, 69% were taking medication for it resulting in an overall medication rate of 6.1%, and increase of 28% from 2007.  About half of children with current ADHD were engaged in some kind of counseling.

There was substantial variation in diagnosis rates and medication treatment between the states with several southern states, such as Arkansas and Kentucky among the highest.  The rate in Vermont was 9.4% which was slightly below the national average and a bit lower than neighboring states (with the exception of New York at 7.7%).  These rates compare to 6.9% in 2003 and 9.9% in 2007 (from the prior CDC report), suggesting that unlike other states, the prevalence in Vermont has not increased from 2007 and may actually be going the other direction.   The 2011 Vermont medication rate was 6.6% which is just above the national average.  Thus, compared to other states, a higher proportion of those diagnosed with ADHD in Vermont are receiving medication treatment.

The authors of the study concluded that there has been a substantial rise in ADHD diagnosis over the past decade with approximately 2 million more children and adolescents receiving the diagnosis now, and most taking medications for it.  This increase, they state, is resulting in an additional burden on the health care system, although the authors don’t do any kind of economic analysis of the costs related to untreated ADHD.

This paper comes soon after a very similar report by McDonald and Jalbert in Psychiatric Times that was previous a topic of this blog.  In that study, the rate of stimulant treatment, using a prescription database, was 2% with Vermont being 34th in the nation.   It is difficult to know exactly how to reconcile these two studies.

While providing very useful information regarding overall statistics, trends, and state differences, it is important to remember that the study has no information regarding the accuracy or appropriateness of the diagnosis or treatment.  Thus, you can expect different people to use these numbers to make different conclusions.   A group here is currently working on some data regarding antipsychotic medications among Vermont prescribers that will shed more light on this important topic.


Visser et al., Trends in the Parent-Report of Health Care Provider-Diagnosed and Medicated Attention-Deficit/Hyperactivity Disorder: United States, 2003–2011. JAACAP 2014;53:34-46.

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