GME Logo

User login

User menu

Issue 68, Dec 2017
Share This
Can ADHD start in adolescence or adulthood?

Sibley MH, Rohde LA, Swanson JM, Hechtman LT, Molina BSG, Mitchell JT, Arnold LE, Caye A, Kennedy TM, Roy A, Stehli A; Multimodal Treatment Study of Children with ADHD (MTA) Cooperative Group. Late-Onset ADHD Reconsidered With Comprehensive Repeated Assessments Between Ages 10 and 25. Am J Psychiatry. 2017 Oct 20:appiajp201717030298. [Epub ahead of print] PubMed PMID: 29050505.


It is fairly common that we see an adolescent or a young adult who presents to us seeking treatment for symptoms of attention deficit/hyperactivity disorder (ADHD) but who does not have a history of ADHD in childhood.

Recently, some studies that enrolled children at an early age and then followed them for many years (called “birth cohort” studies) have suggested that a “late-onset” form of ADHD also exists.

But, the authors of the paper we are discussing here argue that the birth-cohort studies previously done have many limitations. Specifically, they identified ADHD only by screening questionnaires, did not consider alternative causes of the ADHD-like symptoms, and did not obtain complete psychiatric histories.

In this study, the authors have tried to address the limitations of previous studies by looking at more detailed psychiatric assessments administered over time.


This study used the control group from a previously conducted study called the Multimodal Treatment Study of ADHD.

Participants in that control group (N=239) entered the study with no diagnosis of childhood ADHD.

They were then followed up and were assessed eight times over many years.

The mean age at the first, baseline assessment was about 10 years and at the last assessment was done at about 24 years of age.

The diagnostic evaluation used information from parents, teachers, and the subjects.

Impairment due the symptoms, substance use, and other mental disorders were also assessed. The context in which the symptoms occurred and their timing were evaluated.

To optimize sensitivity and specificity, the authors first cast an intentionally wide net for ADHD symptoms to protect against false negatives.

Then, to protect against false positives, they carefully assessed patients, required that meaningful impairment from the symptoms be present for them to be considered relevant, noted whether symptoms were present across different settings, and ruled out substance abuse or other mental disorders as the source of ADHD-like symptoms.


Out of 239 participants who did not have ADHD at baseline:

- 96 met the symptom criteria for ADHD at least one follow up assessment

- Of these 96 participants, 32 also had clinically significant impairment at the time that the ADHD symptoms were present

- Of these 32 participants, 21 had onset of symptoms in adolescence.

- Of these 21 participants, in 3 cases, the symptoms were attributed to cannabis-use disorder.

- That leaves 18 persons. Of these 18 persons, nine had a history of pre-existing or concurrent mental disorders and were reviewed by a clinical panel.

Eight experienced clinicians (three psychiatrists, five clinical psychologists) reviewed the onset and chronicity of all mental symptoms, and each voted as to whether a participant should or should not be considered to have ADHD. This was based on whether the ADHD symptoms or impairment were attributable to another disorder (e.g., effect of anxiety symptoms on concentration). Most decisions were unanimous.

Of the 9 participants with a history of pre-existing or concurrent mental disorders, the panel voted to exclude five based on evidence that symptoms better reflected another mental disorder. Of the remaining 13 (18 minus 5), only six had symptoms in more than one setting.

So, 90 out of 96 (95%) of participants who initially screened positive for ADHD symptoms were not found to have late-onset ADHD based on the more careful evaluation.

The average age at onset in the remaining six patients was about 14 years. Four of these six met symptom criteria only during the teenage years and did not receive any treatment for ADHD

Now, two things have to be noted about these six persons. Two had symptoms of ADHD that persisted into their twenties. Also, five of the six who first met full criteria for ADHD as adolescents had some symptoms of ADHD in childhood (more than most others) even if they did not meet full criteria for ADHD.

In addition to the participants discussed above in detail who had onset of ADHD-like symptoms in adolescence, another 24 participants met the full symptom and impairment criteria for ADHD for the first time after adolescence, i.e., as adults. While details are not discussed here, it should be noted that 14 of them had the impairing ADHD-like symptoms exclusively in the context of heavy substance use.


The authors note that some persons who seek treatment for late-onset ADHD may be valid cases. In a few persons, they found later onset of symptoms meeting full ADHD criteria and no alternative explanation.

But, more commonly, the symptoms suggestive of ADHD are actually due to non-impairing cognitive fluctuations, a comorbid disorder, or the cognitive effects of substance use.

They caution that if careful assessment is not done, a false positive diagnosis of late-onset ADHD can be made.  

They recommend that clinicians should carefully assess impairment, psychiatric history, and substance use before treating persons who present with symptoms suggestive of ADHD but without any childhood history of ADHD.

Clinical Commentary

I don’t think that clinicians are diagnosing their patients with ADHD only based on a screening questionnaire or diagnosing ADHD when the symptoms occur only during heavy substance use. Or, at least, I hope they are not.

But, I strongly agree with the authors’ warning to not diagnose late-onset ADHD without carefully reviewing the context of the symptoms and considering alternative explanations for the symptoms.

I have a lot of clinical experience treating persons with ADHD. I have seen many patients who were not impaired in childhood because they had above average intelligence and a lot of family support. So, as others have suggested, in some patients it is also possible that it is not that the ADHD had late onset but rather it became clinically impairing at a later age.

Expand Content
Share this Section:
First pill with a sensor approved by the FDA

Peters-Strickland T, Pestreich L, Hatch A, Rohatagi S, Baker RA, Docherty JP, Markovtsova L, Raja P, Weiden PJ, Walling DP. Usability of a novel digital medicine system in adults with schizophrenia treated with sensor-embedded tablets of aripiprazole. Neuropsychiatr Dis Treat. 2016 Oct 11;12:2587-2594. eCollection 2016. PubMed PMID: 27785036; PubMed Central PMCID: PMC5067053.

Masand P, Han C, Pae CU. Will the Proteus sensor enhance adherence to aripiprazole or other antipsychotics? Expert Rev Neurother. 2017 Apr;17(4):319-321. PubMed PMID: 28132567.


Non-adherence is a big problem in persons with schizophrenia. Patient self-report and clinician impression have been shown to underestimate non-adherence. Can digital technology help to better monitor adherence?

On November 13, 2017, the United States Food and Drug Administration (FDA) approved a new product containing aripiprazole with a sensor with the brand name Abilify MyCite.

The product contains aripiprazole along with a sensor called an Ingestible Event Marker (IEM). When the pill is ingested by the patient, the sensor sends a signal to a patch that the patient wears on the left lower ribcage. This signal is then transmitted from the sensor on the ribcage to an app on a digital device (either a smartphone or a tablet). Together, this is called a Digital medicine system (DMS). The patient can give permission to caregivers and clinicians to access this information.

The product is approved by the FDA for the same indications as the drug (brand name Abilify).

The paper discussed here evaluated this Digital Management System. The aim was to evaluate the usability and acceptability of the system.


This was a multicenter, phase IIa, open-label study. Patients were not blinded and there was no control group.

Adult patients with schizophrenia were enrolled and treated with aripiprazole with sensor.

After enrollment, patients first entered a training phase consisting of three visits at one-week intervals.

This was followed by a five-week “independent phase”.


Sixty-seven patients with schizophrenia were enrolled in the study and 73% completed the study.

The mean age of the participants was about 47 years. The majority of them were male (75%) and black (76%).

Importantly, 70% of the patients were rated to be mildly ill.

Participants were asked to apply the wearable sensor, change it about once a week, and to pair it each time with the digital device. By the end of the study, 55% of the participants were able to do these tasks independently and another 27% were able to do them with minimal assistance.

The participants wore the wearable sensor for a median of 78% of the time in the trial.

At the last visit, 78% of patients reported being “somewhat satisfied,” “satisfied,” or “extremely satisfied” with the digital management system. But “somewhat satisfied” is a rating that may be easily achieved. Unfortunately, the percentage of patients who reported being “satisfied” or “extremely satisfied” was not reported in the paper.


The authors concluded that a high proportion of the participants were able to use the digital management system and reported satisfaction with it.

They also concluded that these data support the potential utility of the Digital Management System in clinical practice.

Clinical Commentary

On the one hand, it is exciting and interesting to see this innovation in treatment. I am hopeful that in the future, significant improvements and innovations will occur in relation to digital technology that could help monitor adherence.

But with regard to this particular study, we must note that the participants were not particularly ill at the time of the study. Also, they were being paid to participate in the study. Lastly, the study lasted for only eight weeks. Yet, it was not uncommon for patients to not use the wearable sensor or to discontinue participation in the study (27% of those enrolled).

My concern is that the system may work the least in those who need it the most—patients who are more severely ill, who have worse insight, and who are more disorganized.

An Editorial about the potential utility of this sensor technology (Masand et al., 2017) noted several limitations of the study and of the technology:

- It will be difficult for the clinician to convince patients who are paranoid that the Digital Management System is not monitoring them in other ways—their thoughts, actions, etc.

- Managing the privacy issues around the collection and storage of personal health data on digital devices (smartphones, tablets) will be challenging.

- Using the Digital Management System is time-intensive

- The system may help to identify non-adherence, but doesn’t help to assess the reasons for it.

In addition to these concerns, it should also be noted that there is no data to suggest that use of the Digital Management System will achieve its intended goal of improving adherence to the medication.

Expand Content
Share this Section:
Does turning down blue light in smartphones really work?

Heo JY, Kim K, Fava M, Mischoulon D, Papakostas GI, Kim MJ, Kim DJ, Chang KJ, Oh Y, Yu BH, Jeon HJ. Effects of smartphone use with and without blue light at night in healthy adults: A randomized, double-blind, cross-over, placebo-controlled comparison. J Psychiatr Res. 2017 Apr;87:61-70. PubMed PMID: 28017916.


Smartphones deliver light to users through Light Emitting Diode (LED) displays.

In this light, blue light has the most potent effect on sleep and mood. So, it has been suggested that if blue light could be suppressed, the harmful effect of light from smartphone screens could be significantly reduced.

This study, funded by Samsung Display in South Korea, looked at this question.

This is the first controlled study of the effects of exposure of blue light from smartphone LED displays at night on humans.


A randomized, double-blind, cross-over, placebo-controlled design was used.

By “cross over” we mean that the same subjects were exposed to one type of light and then crossed over to being exposed to the other type of light. So, the same subjects received exposure to both types of lights. But the order in which they received one type of light first or the other was random.

The evaluations were done by admitting the subjects to the research unit twice--for three days at a time. 

Two types of smartphones were used. One type had a conventional LED display with the full range of light. The other type had an LED display that has been newly developed by the sponsor of this study, Samsung Display. This type of smartphone suppresses the blue light portion of the spectrum of light.

Was this study blinded? Yes, the two types of smartphones were indistinguishable from each other with the naked eye because the researchers used other wavelengths to mimic blue light in the smartphones in which blue light was suppressed.

Persons with any mental disorder, substance use disorder, or primary sleep disorder, or those who were on any psychotropic medication were excluded from participation in this study.

Each subject played smartphone games with either a conventional LED smartphone or a suppressed blue light smartphone from 7:30 to 10:00 pm. Evaluations were done.

A few days later, they were readmitted, but given the other type of smartphone to play with, and the evaluations were repeated.

The evaluations included changes in serum melatonin levels, cortisol levels, body temperature, and various psychiatric measures of mood, sleep, fatigue. Their scores on the video games they played were also monitored. A Continuous Performance Test was also used.


Twenty-two participants completed both phases of the study.

Use of smartphones without blue light suppression was associated with statistically significantly decreased sleepiness, decreased changes on the Profile of Mood States questionnaire that would be expected with sleepiness, and increased commission errors on the Continuous Performance Test.

Other changes (serum melatonin and cortisol levels) were not statistically significant.


Use at night of LED smartphones without blue light suppression may impair sleep and cognitive performance (increased commission errors).

Clinical Commentary

All patients with sleep-related complaints should be asked to avoid or minimize use of computers, tablets, or smartphones in the few hours prior to bedtime.

But given that these devices have become an indispensible part of our lives, this is easier said than done.

So, we should advise our patients that when use of computer, tablet, or smartphone is unavoidable, blue light suppression should be used.

Expand Content
Share this Section:
A treatment for postmenopausal women with hypoactive sexual desire disorder

Achilli C, Pundir J, Ramanathan P, Sabatini L, Hamoda H, Panay N. Efficacy and safety of transdermal testosterone in postmenopausal women with hypoactive sexual desire disorder: a systematic review and meta-analysis. Fertil Steril. 2017 Feb;107(2):475-482.e15. Review. PubMed PMID: 27916205.


Hypoactive sexual desire disorder (HSDD) is a common condition.

As women age, not only do the estrogen levels decrease, but the androgen levels go down as well.

Previous studies have suggested that transdermal testosterone may be effective as a treatment for hypoactive sexual desire disorder (HSDD) in postmenopausal women.

This paper reports on a meta-analysis of all the previous randomized, placebo-controlled clinical trials on this topic.


Seven short-term, randomized, controlled clinical trials were identified from a thorough search of the literature.

Participants were women who were either naturally or surgically post-menopausal.

There were a total of 3,035 participants in these trials. Of these, 1,350 women were randomized to treatment with a testosterone patch/gel and 1,379 women were randomized to receive either placebo or no treatment.

In most of the studies, the dose of testosterone was 300 mcg per day.

The results were combined using a meta-analytic approach.


 Post-menopausal women who were randomized to receive a testosterone patch/gel had statistically significantly more “satisfying sexual episodes” (the primary outcome measure), sexual activity, orgasms, desire, and improvement in Personal Distress Scale score.

The testosterone group also had more androgenic adverse events like acne and hair growth compared with the control group.

But, there was no statistically significant difference between the two groups in increase in facial hair, alopecia, voice deepening, urinary symptoms, breast pain, headache, site reaction to the patch, serious adverse events, reasons for withdrawal from the study, and the number of women who completed the study.


In these short-term studies, transdermal testosterone was efficacious for the treatment of hypoactive sexual desire disorder (HSDD) in post-menopausal women.

This treatment was associated with androgenic adverse effects like acne and hair growth, but not with more serious adverse effects.

Clinical Commentary

Hypoactive sexual desire disorder (HSDD) in post-menopausal women is a common and distressing problem that is being neglected.

This meta-analysis suggests that we should consider recommending the use of transdermal testosterone in these women. But, as the authors point out, we need more data on the long-term effects of testosterone treatment on breast cancer, coronary heart disease, mood alteration, etc.

Expand Content
Share this Section:
The effects of long-term lithium treatment on the kidney

Tondo L, Abramowicz M, Alda M, Bauer M, Bocchetta A, Bolzani L, Calkin CV, Chillotti C, Hidalgo-Mazzei D, Manchia M, Müller-Oerlinghausen B, Murru A, Perugi G, Pinna M, Quaranta G, Reginaldi D, Reif A, Ritter P Jr, Rybakowski JK, Saiger D, Sani G, Selle V, Stamm T, Vázquez GH, Veeh J, Vieta E, Baldessarini RJ. Long-term lithium treatment in bipolar disorder: effects on glomerular filtration rate and other metabolic parameters. Int J Bipolar Disord. 2017 Dec;5(1):27. PubMed PMID: 28480485; PubMed Central PMCID: PMC5537163.


Clinicians are concerned about the potential adverse effects on the kidney of long-term treatment with lithium.

But, it is not clear how common this is and what the effects of different risk factors are.


Data was collected from 312 persons with bipolar disorder who had been treated clinically at 12 collaborating sites in different countries. These were specialized mood disorders centers and several of the authors are recognized as being leading experts on mood disorders.

Evaluations included changes over time in estimated glomerular filtration rate (eGFR), serum creatinine, urea-nitrogen, serum glucose, white blood cell count, and body-mass index.

The association between estimated glomerular filtration rate (eGFR) and potential risk factors were tested.

An important limitation was the study did not have control data from age-matched subjects who had not taken lithium. The comparison data used in this study were taken from a previous, separate study that had looked at change in eGFR over time.


The patients varied from 20 to 89 years in age (mean 56 years).

They had been treated with lithium carbonate for between 8 to 48 years (mean 18 years).

A low estimated glomerular filtration rate (eGFR) present on more than one occasion (a widely accepted criterion) occurred in 18% of patients.

There was a greater probability of having a low estimated glomerular filtration rate (eGFR) in:

- Persons more than 55 years old, and

- Persons who had taken lithium for 15 years or more

No patient experienced end-stage renal failure.

The decline in eGFR was 19% more in women compared to men.

The mean values of other parameters also increased over time:

- Serum creatinine from 0.87 to 1.17 mg/dL

- BUN from 23.7 to 33.1 mg/dL

- Glucose from 88 to 122 mg/dL, and

- BMI from 25.9 to 26.6 kg/m(2).

Using multivariate regression, the most important risk factors for a decline in the eGFR were:

- Longer duration of lithium treatment

- Lower (sic) lithium dose (without a lower serum lithium concentration). Note: The dose of lithium carbonate decreased over the years even though serum lithium concentrations remained the same. This is presumably because the doses were adjusted as lithium clearance decreased with increasing age.

- Older age

- Medical comorbidity

- Lower initial eGFR, and

- Starting lithium at age 40 years or more.

The decline in eGFR was noted starting from years 6 to 10. The authors projected that if the trend continued, a decline in eGFR to the lower limit of normal would occur after 30 or more years of taking lithium.


Long-term lithium treatment was associated with gradual decline in renal functioning (eGFR) to a greater extent than by age alone, at least based on comparison with a control group from a previous study.  

Various risk factors for decline in renal function were identified.

Clinical Commentary

In the past, many clinicians and academicians did not accept that there was a risk of decline in renal function with long-term lithium treatment. But in the last few years, even though the studies are sometimes contradictory, data have accumulated to show that slow decline in eGFR occurs, especially after 15 or more years of treatment.

But, better data with larger samples, age-matched controls, and a certain minimum number of years of treatment with lithium are needed.

In the meanwhile, we must continue to pursue informed consent and regular monitoring for patients who are prescribed lithium.

We should not be reassured that in this study no participant developed end-stage renal disease. Uncommon outcomes cannot be identified in relatively small samples. Also, as the authors noted, the participants were being treated in specialized mood-disorder clinics where close clinical follow-up would have led to discontinuation of lithium before a patient progressed to end-stage renal disease.

Expand Content
Share this Section:

Rajnish Mago, MD
Medical Editor, GME Research Review

GME Research Review is a monthly newsletter edited by Rajnish Mago, MD, who is author of "The Latest Antidepressants" and "Side Effects of Psychiatric Medications: Prevention, Assessment, and Management." Dr. Mago selects, summarizes, and provides a clinical commentary on the latest published research in psychiatry. 

We are always carefully evaluating which research papers to discuss in GME Research Review. Have come across a research paper published in the last 6 months that you thought is clinically relevant? Do you want me to analyze it for you and for the benefit of others? Please email Dr. Mago the citation at [email protected].

To contact GME, email us at [email protected]

GME does not provide medical advice. The website and articles are intended for informational purposes only. They are not a substitute for professional medical advice, diagnosis or treatment. Never ignore professional medical advice in seeking treatment because of something you have read on the GME Website. If you think you may have a medical emergency, immediately call your doctor or dial 911.

Log In

Join GME For Free

Become a GME subscriber and gain full access to our extensive library of 700+ psychiatric medical education videos, free CME webcasts, latest research updates, and more. To sample our content, watch the featured videos below.
Join Today!