Stephen V. Faraone, PhDWe are only beginning to explore how ADHD affects sleep in adults. A team of European researchers recently published the first meta-analysis on the subject, drawing on thirteen studies with 1,439 participants. They examined both subjective evaluations from sleep questionnaires and objective measurements from actigraphy and polysomnography. However, due to differences among the studies, only two to seven could be combined for any single topic, generally with considerably fewer participants (88 to 873).

Several patterns emerged. Looking at results from sleep questionnaires, they found that adults with ADHD were far more likely to report general sleep problems (very large SMD effect size 1.55). Getting more specific, they were also more likely to report frequent night awakenings (medium effect size 0.56), taking longer to get to sleep (medium-to-large effect size 0.67), lower sleep quality (medium-to-large effect size 0.69), lower sleep efficiency (medium effect size 0.55), and feeling sleepy during the daytime (large effect size 0.75). There was little to no sign of publication bias, though considerable heterogeneity on all but night awakenings and sleep quality.

Actigraphy readings confirmed some of the subjective reports. On average, adults with ADHD took longer to get to sleep (large effect size 0.80) and had lower sleep efficiency (medium-to-large effect size 0.68). They also spent more time awake (small-to-medium effect size 0.40). There was little to no sign of publication bias and there was little heterogeneity among studies.

None of the polysomnographic measurements, however, found any significant differences between adults with and without ADHD. All effect sizes were small (under 0.20), and none came close to being statistically significant.

There were four instances where measurement criteria overlapped those from actigraphy and self-reporting, with varying degrees of agreement and divergence. There was no significant difference in total sleep time, matching findings from both the questionnaires and actigraphy. On percent time spent awake, polysomnography found little to no effect size with no statistical significance, whereas actigraphy found a small-to-medium effect size that did not quite reach significance, and self-reporting came up with a medium effect size that was statistically significant. On sleep onset latency and sleep efficiency, for which questionnaires and actigraphy found medium-to-large effects, the polysomnographic measurements found little to none, with no statistical significance.

Polysomnography found no significant differences in stage 1 sleep, stage 2 sleep, slow wave sleep, and REM sleep. With the exception of slow wave sleep, there was no sign of publication bias. Heterogeneity was generally minimal.

One problem with the extant literature is that many studies did not take medication status into account. In fact, the authors concluded, “future studies should be conducted in medication naïve samples of adults with and without ADHD matched for comorbid psychiatric disorders and other relevant demographic variables.”

In summary, these findings provide robust evidence that ADHD adults report a variety of sleep problems. In contrast, objective demonstrations of sleep abnormalities have not been consistently demonstrated. More work in medication naïve samples is needed to confirm these conclusions.

REFERENCES
Amparo Díaz-Román, Raziya Mitchell, Samuele Cortese, “Sleep in adults with ADHD: Systematic review and meta-analysis of subjective and objective studies,” Neuroscience and Biobehavioral Reviews, vol. 89, p. 61-71 (2018).

Stephen V. Faraone, PhDA systematic review of the literature found seven studies examining this question. Significantly, six were large cohort studies with a combined total of almost three million individuals. The other was a large case-control study with 7,874 participants.

The largest cohort study, with more than a million and a half children, found that prenatal antidepressant exposure increased the risk for ADHD. The adjusted odds ratio was 1.6 for any antidepressant and for selective serotonin reuptake inhibitors (SSRI). But in sibling comparison models, which better adjust for confounds shared by siblings (e.g., poverty, stress in the home), this study found no increased risk of ADHD.

The second largest cohort study, with over 875 thousand children, found a small adjusted risk of 1.2 for all antidepressants, with little variation by class of antidepressant. The fourth largest study, with over 140 thousand children, likewise found a small adjusted risk of 1.2, which barely achieved statistical significance (95% CI 1.0-1.4).

The third largest study, with over 190 thousand children, obtained an adjusted risk of 1.4 for all antidepressants. But it also pointed to a possible explanation for the small association found in this and other studies suggesting that the apparent association with antidepressant use was due to ADHD’s known genetic association with psychiatric conditions treated by antidepressants.

The fifth largest study, with more than 55 thousand children, similarly found an adjusted risk of 1.7 for SSRIs and an adjusted risk of 1.7 for unmedicated maternal psychiatric disorder. Again, the underlying psychiatric disorder appears to be confounding the effect of antidepressants.

The sixth largest study, with over 38 thousand children, found no evidence of any effect from SSRIs. Yet it found evidence of a large effect from bupropion, with an odds ratio of 3.6, and only one in 50 odds of obtaining such a result by chance (p = 0.02). However, it offered no comparison with untreated depression, and made no adjustments for potential confounders.

The case control study found an odds ratio of 2.3 for maternal use of any antidepressant, which dropped to a statistically nonsignificant 1.6 when adjusted for maternal psychiatric disorder (95% CI 0.66-3.71).

The review concludes, “The evidence available is inadequate to indicate any negative effects of a specific class of antidepressant on the risk of ADHD.”

REFERENCES
Faruk Uguz, “Maternal Antidepressant Use During Pregnancy and the Risk of Attention-Deficit/Hyperactivity Disorder in Children: A Systematic Review of the Current Literature,” Journal of Clinical Psychopharmacology, vol. 38, no. 3 (2018).

Stephen V. Faraone, PhDFew studies have examined the safety and tolerability of ADHD medications (stimulants and atomoxetine) extending beyond six months, and none beyond a few years. A pair of Swedish neuroscientists at Uppsala University Hospital set out to explore longer-term outcomes. They conducted a six-year prospective study of 112 adults diagnosed with ADHD who were being treated with ADHD medications (primarily MPH, but also dexamphetamine and atomoxetine).

They found that at the end of that period, roughly half were still on medication, and half had discontinued treatment. There were no significant differences between the two groups in age, sex, ADHD severity, or comorbidity. The average ADHD score for the entire cohort declined very significantly, from a mean of 37 to a mean of 26, with a less than one in a thousand odds of that being due to chance. There was also no sign of drug tolerance or of a need to increase dosage over time.

All 55 adults who discontinued treatment had taken MPH for at least part of the time. Eleven had also been treated with dexamphetamine (DEX) and 15 with atomoxetine (ATX). The average time on treatment was just under two years. Almost a third quit MPH because they perceived no beneficial effect. Since they were on average taking higher doses at discontinuation than at initiation, that is unlikely to have been due to suboptimal dosage. Almost another third discontinued for various adverse mental effects, including hyperactivity, elation, depressive moods, aggression, insomnia, fatigue, and lethargy. Another one in eleven quit when they lost contact with the prescribing physician. In the case of ATX, almost half quit because of what they perceived as adverse mental effects.

Among the 57 adults who remained on medication, four out of five reported a strong beneficial effect. Only two reported minimal or no effect. Compared with the group that discontinued, the group that remained on medication was far more likely to agree with the statements, “My quality of life has improved,” and “My level of functioning has improved.” Yet as the authors caution, it is possible “that the subjects’ subjective ratings contain a placebo-related mechanism in those who are compliant with the medication and pursue treatment over time.” In fact, the authors reported that there were no significant differences in ADHD scores or ADHD severity between the group that quit and the group that remained on medication, even though, on average, the group that quit had been off medication for four years at follow-up.

We cannot explain why the patients who quit treatment showed similar levels of ADHD symptoms to those who continued treatment. It is possible that some patients remit symptoms over time and do not require sustained treatment. But we must keep in mind that there was a wide range of outcomes in both groups. Future work needs to find predictors of those who will do well after treatment withdrawal and those who do not.

Any decision on whether to maintain a course of medication should always weigh expected gains against adverse side effects. Short of hard evidence of continuing efficacy beyond two years, adverse events gain in relative importance. With that in mind, it is worth noting that this study reports that among those who remained on MPH, many reported side effects. More than a quarter complained of decreased appetite, one in four of dry mouth, one in five of anxiousness and of increased heart rate, one in six of decreased sexual desire, one in nine of depressed mood, and one in eleven of insomnia.

This study breaks important ground in looking at long-term effects of medication. It reaffirms findings elsewhere of the efficacy of ADHD medications. But contrary to the authors’ conclusion, the data they present suggests the possibility that permanently medicating ADHD patients may not be more efficacious than discontinuation beyond a certain point, especially when balanced against adverse side effects.

But this is just one study with a relatively small sample size. This suggests a need for additional studies with larger sample sizes to pursue this question with greater statistical reliability.

REFERENCES
Dan Edvinsson and Lisa Ekselius, “Long-Term Tolerability and Safety of Pharmacological Treatment of Adult Attention-Deficit/Hyperactivity Disorder,” Journal of Clinical Psychopharmacology, vol. 38, no. 4 (2018).

Stephen V. Faraone, PhDAn international group of twelve experts recently published a consensus report examining the state of the evidence and offering recommendations to guide screening, diagnosis, and treatment of individuals with ADHD-SUD comorbidity.[1]

In a clear sign that we are still in the early stages of understanding this relationship, five of the thirteen recommendations received the lowest recommendation grade (D), eight received the next-lowest (C), and none received the highest (A and B). The lower grades reflected the absence of the highest level of evidence, obtained from meta-analyses or systematic reviews of relevant randomized controlled trials (RCTs).

Nevertheless, with these limitations in mind, the experts agreed on the following points:

Diagnosis

  • The strongest recommendation, the only one based on a 2+ level of evidence (well-conducted case control or cohort studies with a low risk of confounding or bias and a moderate probability that the relationship is causal) is that the “Short Version of the Adult ADHD Self-Report Scale (ASRS-SV) screener is currently the most widely used and investigated screening tool in individuals with ADHD and comorbid SUD, with good sensitivity and specificity across studies.”
  • Two other recommendations were graded C: The diagnostic process should include current and past substance abuse and seek to involve partners and relatives in evaluating symptoms and functional impairments.
  • Four recommendations got the lowest grade, D. The experts suggested starting the diagnostic process as soon as possible and focusing on drug- and alcohol-free periods in the patient’s life during history taking. They also recommended that physicians and clinical psychologists should only make diagnoses if they have extensive training in diagnosing ADHD, as well as experience with adults with ADHD and with addiction care, and that they should consider treating adults with sufficiently severe ADHD symptoms.

Treatment

  • In general, evidence was stronger in this area, and only one of the six recommendations was graded D. The other five recommendations were graded C, with the highest level of evidence being 2 (cohort or case and control studies with undetermined risk of bias), although in three cases it was level 3 (non-analytical studies, such as case reports and case series).
  • The grade D recommendation was to always consider a combination of psychotherapy and pharmacotherapy.

The grade C recommendations included considering adequate medical treatment of both ADHD and SUD; integrating ADHD treatment with SUD treatment as soon as possible;
considering psychotherapy targeting both; use of long-acting methylphenidate, extended-release amphetamines, and atomoxetine because of their low potential for abuse; and careful clinical management to avoid abuse and diversion of prescribed stimulants.

[1] Cleo L. Crunelle at al., “International Consensus Statement on Screening, Diagnosis and Treatment of Substance Use Disorder Patients with Comorbid Attention Deficit/Hyperactivity Disorder,” European Addiction Research, published online March 6, 2018, DOI: 10.1159/000487767.

Stephen V. Faraone, PhDA team of U.S. endocrinologists recently published the results of a meta-analysis examining a possible association between bisphenol A (BPA) and childhood ADHD. BPA is used in a variety of consumer products, including plastic bottles for food and drink, epoxy resins used to line cans of food, dental sealants, and the thermal receipts issued by stores.

A review of the literature found 29 rodent studies but only three with humans. The human studies were too different from each other to be suitable for meta-analysis. One found no association between prenatal exposure and ADHD. A second found prenatal BPA exposure to be associated with teacher-reported hyperactivity in 4-year-old boys, but not girls. The third found it to be associated with hyperactivity scores in 3-year-old girls.

As the authors note, “Often, there is little human data available, particularly in the environmental toxicology/health fields, due to the time and expense of conducting epidemiological studies and the ethical barriers for human controlled trials that involve human exposure to potentially hazardous chemicals. Thus, it is important to have methods for using animal data to inform human health hazard conclusions; indeed, animal models are traditionally used to study human health.”

Twelve of the mice and rat studies, with a total of 709 rodents, were suitable for meta-analysis.

Overall these pointed to a tiny SMD effect size of 0.09, but it was not significant, with the odds of such a result being obtained by chance being almost one in four (p = 0.237). But when results from the 356 males and 353 females were looked at separately, a significant sex difference emerged. There was essentially no effect on female rodents, with an effect size of -0.07 and a 95% confidence interval of -0.27 to 0.14, widely spanning the zero mark, rendering the result statistically nonsignificant. Among male rodents, however, there was a small but statistically significant effect size (0.24), with a 95% confidence interval from 0.04 to 0.45. The odds of obtaining this outcome by chance were only one in 50 (p = .02).

This result must be viewed with caution, as rodent physiology often differs substantially from that of humans. The authors therefore conclude, “early BPA exposure is associated with a presumed hazard of hyperactivity in humans. Our conclusion is based on ‘moderate’ levels of evidence for the human and ‘high’ levels of evidence for animal literature.”

REFERENCES
Johanna R. Rochester, Ashley L. Bolden, Carol F. Kwiatkowski, “Prenatal exposure to bisphenol A and hyperactivity in children: a systematic review and meta-analysis,” Environment International, vol. 114, p. 343-356 (2018).

ADHD Affects the Efficacy of Treatment for Eating Disorders in Adult Women

Stephen V. Faraone, PhDSwedish researchers examined outcomes for adult women who sought treatment at the Stockholm Centre for Eating Disorders over a period of two years and nine months. Out of 1,517 women who came to the clinic 1,143 remained eligible for the study, after excluding women whose symptoms did not fulfill the DSM-IV criteria for eating disorders or had incomplete records.

Of these, seven hundred patients could not be reached or declined to participate, leaving 443 for follow-up. To guard against the possibility that the follow-up group might not be representative of the overall treatment group, researchers compared age, body mass index, and scores on tests for depression, anxiety, compulsivity, inattention, and hyperactivity. The only statistically significant differences were small ones. The median age of the group lost to follow-up was one year younger, they were less likely to be living alone, and on average scored a single point higher on the depression test. Otherwise they were broadly similar.

The one-year follow-up on the study group found a substantial difference in rate of recovery from eating disorders between those wEating disorders and ADHDith and without comorbid ADHD. Almost three out of four patients (72%) who scored lower (between 0-17) on the World Health Organization adult ADHD self-report scale had recovered from their eating disorder. Among those scoring 18 and higher, on the other hand, it was less than half (47%). This difference was extraordinarily unlikely (one chance in one thousand) to be due to chance (p=.001).

Another way of expressing this is through odds ratios. Those scoring 18 and up on the ADHD self-report scale were about two and a half times less likely to recover from their eating disorders following treatment. More specifically, they were about three times less likely to recover from loss of control and binging, and almost three and a half times less likely to recover from purging.

To improve outcomes, the researchers suggest “identifying concomitant ADHD symptoms and customizing treatment interventions based on this.” They specifically propose controlled clinical trials to explore the effect of combining stimulant medications with standard treatment for eating disorders.

REFERENCES
Nils Erik Svedlund, Claes Norring, Ylva Ginsberg, Yvonne von Hausswolff‐Juhlin, “Are treatment results for eating disorders affected by ADHD symptoms? A one‐year follow‐up of adult females,” European Eating Disorders Review (2018).

Stephen V. Faraone, PhDAn international group of twelve experts recently published a consensus report examining the state of the evidence and offering recommendations to guide screening, diagnosis, and treatment of individuals with ADHD-SUD comorbidity.1

In a clear sign that we are still in the early stages of understanding this relationship, five of the thirteen recommendations received the lowest recommendation grade (D), eight received the next-lowest (C), and none received the highest (A and B).

The lower grades reflected the absence of the highest level of evidence, obtained from meta-analyses or systematic reviews of relevant randomized controlled trials (RCTs).

Nevertheless, with these limitations in mind, the experts agreed on the following points:

ADHD Diagnosis

  • The strongest recommendation, the only one based on a 2+ level of evidence (well-conducted case control or cohort studies with a low risk of confounding or bias and a moderate probability that the relationship is causal) is that the “Short Version of the Adult ADHD Self-Report Scale (ASRS-SV) screener is currently the most widely used and investigated screening tool in individuals with ADHD and comorbid SUD, with good sensitivity and specificity across studies.”
  • Two other recommendations were graded C: The diagnostic process should include current and past substance abuse and seek to involve partners and relatives in evaluating symptoms and functional impairments.
  • Four recommendations got the lowest grade, D. The experts suggested starting the diagnostic process as soon as possible and focusing on drug- and alcohol-free periods in the patient’s life during history taking. They also recommended that physicians and clinical psychologists should only make diagnoses if they have extensive training in diagnosing ADHD, as well as experience with adults with ADHD and with addiction care, and that they should consider treating adults with sufficiently severe ADHD symptoms.

ADHD Treatment

  • In general, evidence was stronger in this area, and only one of the six recommendations was graded D. The other five recommendations were graded C, with the highest level of evidence being 2 (cohort or case and control studies with undetermined risk of bias), although in three cases it was level 3 (non-analytical studies, such as case reports and case series).
  • The grade D recommendation was to always consider a combination of psychotherapy and pharmacotherapy.
  • The grade C recommendations included considering adequate medical treatment of both ADHD and SUD; integrating ADHD treatment with SUD treatment as soon as possible; considering psychotherapy targeting both; use of long-acting methylphenidate, extended-release amphetamines, and atomoxetine because of their low potential for abuse; and careful clinical management to avoid abuse and diversion of prescribed stimulants.

Note: Andrew Reding is a co-author on this post.

REFERENCES
1Cleo L. Crunelle at al., “International Consensus Statement on Screening, Diagnosis and Treatment of Substance Use Disorder Patients with Comorbid Attention Deficit/Hyperactivity Disorder,” European Addiction Research, published online March 6, 2018, DOI: 10.1159/000487767.

Stephen V. Faraone, PhDA Dutch study compared the efficacy of mindfulness-based cognitive therapy (MBCT) combined with treatment as usual (TAU), with TAU-only as the control group. MBCT consisted of an eight-week group therapy consisting of mindfulness exercises (bodyscan, sitting meditation, mindful movement), psychoeducation about ADHD, and group exercises. TAU consisted of usual treatment in the Netherlands, including medications and other psychological treatment. Sixty individuals were randomly assigned to each group. MBCT was taught in subgroups of 8 to 12 individuals. Patients assigned to TAU were not brought together in small groups. Baseline demographic and clinical characteristics were closely matched for both groups.

Mindfulness Cognitive Behavioral TherapyOutcomes were evaluated at the start, immediately following treatment, and again after 3 and 6 months using well-validated rating scales. Following treatment, the MBCT + TAU group outperformed the TAU group by an average of 3.4 points on the Conners’ Adult Rating Scale, corresponding to a standardized mean difference of .41. Thirty-one percent of the MBCT + TAU group made significant gains, versus 5% of the TAU group. 27% of MBCT +TAU patients scored a symptom reduction of at least 30 percent, as opposed to only 4% of TAU patients. Three and six-month follow-up effects were stable, with an effect size of .43.

The authors concluded “that MBCT has significant benefits to adults with ADHD up to 6 months after post-treatment, with regard to both ADHD symptoms and positive outcomes.” Yet in their section on limitations, they overlook a potentially important one. There was no active placebo control. Those who were undergoing TAU-only were aware that they were not doing anything different from what they had been doing before the study. Hence no substantial placebo response would be expected from this group during the intervention period (post-treatment they were offered an opportunity to undergo MBCT). Moreover, MBCT + TAU participants were gathered into small groups, whereas TAU participants were not. We therefore have no way of knowing what effect group interaction had on the outcomes, because it was not controlled for. So, although these results are intriguing and suggest that further research is worthwhile, the work is not sufficiently rigorous to definitively conclude that MBCT should be prescribed for adults with ADHD.

Note: This post was co-authored by Andrew Reding.

REFERENCES
Janssen L, Kan CC, Carpentier PJ, Sizoo B, Hepark S, Schellekens MPJ, Donders ART, Buitelaar JK, Speckens AEM. “Mindfulness-based cognitive therapy v. treatment as usual in adults with ADHD: a multicentre, single-blind, randomised controlled trial,” Psychological Medicine (2018), https:// doi.org/10.1017/S0033291718000429

Stephen V. Faraone, PhDThough there have been numerous studies of the efficacy of cognitive behavioral therapy (CBT) for ADHD symptoms in children, adolescents, and adults, few have examined efficacy among adults over 50. A new study begins to fill that void.

Psychiatric researchers from the New York University School of Medicine, Massachusetts General Hospital, and Pfizer randomly assigned 88 adults diagnosed with elevated levels of ADHD to one of two groups. The first group received 12 weeks of CBT targeting executive dysfunction – a deficiency in the ability to properly analyze, plan, organize, schedule, and complete tasks. The second group was assigned to a support group, intended to serve as a control for any effects arising from participating in a group therapy. Each group was split into subgroups of six to eight participants. One of the CBT subgroups was run concurrently with one of the support-only subgroups and matched on the percent receiving ADHD medications.

Outcomes were obtained for different ADHD demographics, 26 adults aged 50 or older (12 in CBT and 14 in support) and compared with 55 younger adults (29 in CBT and 26 in support). The mean age of the younger group was 35 and of the older group 56. Roughly half of the older group, and 3/5ths of the younger group, was on medication. Independent (“blinded”) clinicians rated symptoms of ADHD before and after treatment.

In the blind structured interview, both inattentive scores and executive function scores improved significantly and almost identically for both older and younger adults following CBT. When compared with the controls (support groups), however, there was a marked divergence. In younger adults, CBT groups significantlyIs Cognitive Behavior Therapy Effective for Older Adults with ADHD? outperformed support groups, with mean relative score improvements of 3.7 for inattentive symptoms and 2.9 for executive functioning. In older adults, however, the relative score improvements were only 1.1 and 0.9, and were not statistically significant.

Given the nonsignificant improvements over placebo, the authors’ conclusion that “The results provide preliminary evidence that CBT is an effective intervention for older adults with ADHD” is premature. As they note, a similar large placebo effect was seen in adults over 50 in a meta-analysis of CBT for depression, rendering the outcomes nonsignificant. Perhaps structured human contact is the key ingredient in this age group. It may also be, as suggested by the positive relative gains on six of seven measures, that CBT has a small net benefit over placebo, which cannot be validated with such a small sample size. Awaiting results from studies with larger sample sizes, it is for now impossible to reach any definitive conclusions about the efficacy of CBT for treating adults over 50.

Note: Andrew Reding is co-author on this post.

REFERENCES
Mary V. Solanto, Craig B. Surman, Jose Ma. J. Alvir, “The efficacy of cognitive–behavioral therapy for older adults with ADHD: a randomized controlled trial,” ADHD Attention Deficit and Hyperactivity Disorders (2018)

Stephen V Faraone PhD AIA 2016 tatAj0

A recent CNN report, http://tinyurl.com/yannlfd6, highlighted a paper published in Pediatrics, which reported that pregnant women who use acetaminophen during pregnancy put their unborn child at two-fold increased risk for attention deficit hyperactivity disorder (ADHD).   In that study, acetaminophen use during pregnancy was common; nearly half of women surveyed used the painkiller during a pregnancy.   Other studies have reported similar associations of acetaminophen, also known as paracetamol with ADHD or with other problems in childhood (e.g., https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5300094/, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4177119/, https://www.ncbi.nlm.nih.gov/pubmed/24566677, https://www.ncbi.nlm.nih.gov/pubmed/24163279). Given these prior findings, it seems unlikely that the new report is a chance finding. But does it make any biological sense?   One answer to that question came from an epigenetic study. Such studies figure out if assaults from the environment change the genetic code. One epigenetic study found that prenatal exposure, changes the fetal genome via a process called methylation. Such genomic changes could increase risk for ADHD (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5540511/)/ . Because all of these studies are observational studies, one cannot assert with certainty that there is a causal link between acetaminophen use during pregnancy. The observed association could be due to some unmeasured third factor. Although the researchers did a respectable job ruling out some third factors, we must acknowledge some uncertainty in the finding. That said, what should pregnant women do if they need a acetaminophen.   I suggest you bring this information to your physician and ask if there is a suitable alternative.