Does Early Nonreponse to Medication Predict Long-Term Response?
Kathryn G. Fort
Fellow, Child and Adolescent Psychiatry
NYU Child Study Center
Bellevue Hospital Center
Dr. Fort has disclosed that she has no relevant financial or other interests in any commercial companies pertaining to this educational activity.
“My child hasn’t shown improvement since starting this medication over a month ago! Can we still expect it to work?” We’ve all been asked this difficult question, often by desperate parents—especially those of children taking antipsychotics. Now a recent study provides initial data to answer it.
To determine if early nonresponse to second generation antipsychotics was predictive of “ultimate,” or long-term, non-response, data were collected from the SATIETY (Second-Generation Antipsychotic Treatment Indications, Effectiveness and Tolerability in Youth) cohort study. Subjects were selected by chart review, and those included in the analysis were ages four to 19 and prescribed antipsychotics for the diagnoses of psychotic spectrum disorder. Naturalistic treatment was pursued, with the outpatient clinician choosing a medication course as clinically indicated in these primarily antipsychotic-naive subjects (77.2%).
Assessment was conducted with the Children’s Global Assessment Scale (CGAS) and the Clinical Global Impression-Severity (CGI-S) at baseline, with the CGI-Improvement (CGI-I) added monthly in follow-up. Subjects were additionally monitored for side effects, particularly EPS, akathisia, and weight change. Treatment adherence was confirmed by monthly antipsychotic blood and plasma levels.
Seventy-nine children were included in the final analysis. Of note, Early Response (ER) was defined as scores of <3 (at least minimally improved), and Early Nonresponse (ENR) as >4 (worse or no change) on the CGI-I at four weeks. Ultimate Response (UR) (ie, long-term response) was defined by a score of 1 or 2 (much or very much improved), and Ultimate Nonresponse (ie, long-term nonresponse) was >3 (minimally improved or worse) on the GCI-I at eight to 12 weeks.
At week four, 36 subjects (45.6%) were early responders and 43 (54.4%) were early nonresponders. At week 12, 45 subjects (57%) were ultimate responders (UR) and 34 (43%) were ultimate non-responders (UNR).
The primary outcome of the study— the predictive value of early response or early nonresponse for ultimate response or ultimate nonresponse—had a negative predictive value of 67.4% and a positive predictive value of 86.1%. In other words, early response subjects showed greater long-term improvements in clinical ratings by the CGI-I and CGAS than the early nonresponse subjects.
Study authors report these findings with cautious optimism due to concern for small sample size and lack of controlled settings. However, given the approximation to adult and other first- break findings and the use of clinical scales, the clinical predictive value and ease of generalizability of this study positively add to the available literature in this area (Stentebjerg-Olesen M et al, J Child Adolesc Psychopharmacology 2013;23:1-11).
CCPR’s Take: The clinical usefulness of this data is promising; however it will require further investigation and replication with increased controls for further improvement of the knowledge base. It should be reiterated that this was a small study using a chart review of SATIETY participants and that the SATIETY study was not designed to look at this data in particular.
What Meds Work for Insomnia in ADHD?
Sleep complaints are common in children with ADHD, whether they are caused by the disorder itself, the treatments for it, or both. Despite this fact, there are limited studies on treatments for insomnia and other sleep problems in children and adolescents with ADHD.
A recent meta-analysis looked at the research on medications to treat insomnia comorbid with ADHD. Researchers included five studies looking at a total of four medications for a variety of sleep-related factors, including sleep onset, sleep efficiency, and total sleep time.
Included for consideration in the meta-analysis were studies primarily focused on pharmacological treatments for sleep disorders among patients diagnosed with ADHD. Some studies included participants with comorbid mental illnesses, as long as the study’s main objective was analyzing treatment for the sleep disorder and not another psychiatric condition.
In the studies picked for final analysis, participants were both genders and ranged in age from four to 17. Some were on no medications except for those used in the study, others were on medications including mood stabilizers, stimulants, antidepressants, and antihistamines.
The five studies that were ultimately chosen for analysis included a retrospective chart review of clonidine (Catapres, Kapvay) (n=62); a double-blind, placebo-controlled study of zolpidem (Ambien) (n=201); a randomized, double-blind, placebo- controlled study of melatonin (n=105); a case-controlled follow-up study of an RCT of melatonin (n=94); and a double-blind, placebo-controlled study of l-theanine (n=93).
Among the studied medications, melatonin performed best—reducing the time to fall asleep and increasing total sleep time and sleep efficiency. In addition, parents of participants reported significant improvement of core symptoms of ADHD.
L-theanine showed positive results in sleep efficiency (the percent of the night spent asleep), and clonidine reduced insomnia, as measured by the Clinical Global Improvement-Severity (CGI-S) scale.
Zolpidem showed poor results in reducing sleep latency, or in measures of time spent asleep. However, on a subjective measure of the severity of insomnia (CGI-S), both children and their parents reported improvement. More than 62% of children taking zolpidem experienced side effects such as dizziness, headaches, or hallucinations, and more than 7% dropped out.
Melatonin dosages studied ranged from 0.5 to 10 mg; clonidine from 50 to 800 mcg; zolpidem 0.25 mg/kg (max 10 mg); and l-theanine 400 mg daily (Barrett JR et al, J Child Adolesc Psychopharmacology 2013;23(10):1–8).
CCPR’s Take: This study found a non-prescription drug—melatonin—to be effective for a number of sleep complaints in children with ADHD. And while participants in some of the studies were taking other medications, including some that cause sedation on their own, none of the participants in the two melatonin studies were on other drugs, so it’s unlikely any other substance helped with these sleep problems. This analysis did have several limitations, however. The aforementioned possibility that something else participants were taking worked with, against, or in place of the study drug needs to be considered. In addition, the study didn’t include analysis of any non-medication treatments or of combinations of medication and psychotherapy or behavioral modification.