Substance abuse with self-medication effects in ADHD
On closer inspection, some misuse of narcotic substances turns out to be self-medication. This does not change the abusiveness or even the possible criminal liability. In addition, the desired effect of normalization is not achieved and considerable unpleasant side effects occur. However, the tendency to prefer certain drugs can provide information about the causes of substance abuse.
ADHD is characterized, among other things, by an increased number of DAT. One study reports a strong reduction of DAT in ADHD sufferers if they were also users of dopaminergic drugs.1
- 1. Nicotine
2. Cannabis: THC, Cannabidiol
2.1. Cannabinoids (THC)
- 2.1.1. Cannabinoids affect dopamine, serotonin, and acetylcholine in the PFC
- 2.1.2. Increase in the level of cortisol
- 2.1.3. HPA axis shutdown by THC-mediated cortisol increase?
- 2.1.4. Decreased cortisol response to acute stressors
- 2.1.5. Reduction of the amygdala volume
- 2.1.6. Cannabinoids change the EEG: Alpha increased, beta decreased
- 2.1.7. Cannabis abuse correlates with ADHD symptoms, not cognitive problems
- 2.2. Cannabidiol (CBD)
- 2.3. THC drugs for ADHD
- 2.4. ADHD symptom improvement through microdosing (microdosing)
- 2.1. Cannabinoids (THC)
- 3. Amphetamines / Mesamphetamines
- 4. Cocaine
- 5. Alcohol
- 6. LSD
Nicotine and methylphenidates are both stimulants. They have similar mechanisms of action.2 Nicotine as well as methylphenidate reduce the number of dopamine transporters (DAT)3 in the medium term and increase - also thereby - the dopamine availability in the synaptic cleft.4
Empirical evidence shows that non-medicated ADHD-HI sufferers are significantly more likely to smoke than medicated ADHD-HI sufferers or non-smokers.5
The increase in risk for ADHD due to maternal smoking during pregnancy (for more on this topic, see ⇒ Nicotine consumption of the mother during pregnancy In the article ⇒ Environmental factors as a cause of ADHD in the chapter *⇒ Emergence is likely due, at least in part, to the increased likelihood of passing on the genetic disposition that made the mother herself an ADHD sufferer. In addition, the fact that a woman cannot stop smoking despite pregnancy indicates, first, that nicotine is very much needed, second, a lack of self-control, and third, a low level of care toward the entrusted child. Also with regard to the fact that this is more common in low socioeconomic strata than in more affluent strata, the question arises whether belonging to the social stratum is the consequence of impaired self-regulation or vice versa. In any case, a low socioeconomic status correlates with an increased risk for ADHD.
- Nicotine significantly increases adrenaline levels and tends to decrease norepinephrine levels. Nicotine thus improves test results and reaction times.678
- Nicotine patches have shown a positive effect in ADHD in a double-blind study.91011
- Nicotine, or smoking, is reported to significantly increase the risk for depression and to significantly worsen mood.12 Psylex goes on to cite a study that showed a clear causality of smoking for depression, but not depression for smoking. Unfortunately, this source was not verifiable.
- Smoking significantly reduces the number of dopamine transporters in the striatum. An increased number of dopamine transporters is thought to be a typical phenomenon in ADHD. In a SPECT study of 31 adults with ADHD, a greater increase in DAT was found in ADHD-HI subjects than in ADHD-I subjects. However, DAT were still elevated in ADHD-I sufferers compared with nonaffected individuals. Smoking significantly decreased DAT in both subtypes.13
- A study of ADHD-affected smokers found that mild ADHD-affected smokers showed decreased inhibition and increased impulsivity after smoking cessation, whereas this was surprisingly not the case for severe ADHD-affected smokers.14
2. Cannabis: THC, Cannabidiol
Among ADHD sufferers, abuse of hashish/cannabis/marijuana is very common.
One study found cannabis use in 14.3% of all ADHD sufferers, while only 4.3% of non-affected people use cannabis. ADHD thus increases the likelihood of cannabis use more than threefold.15 Hyperactive persons also started their cannabis use much earlier than purely inattentive persons.
The assumption that this is due to the relaxing to sedating effect is questionable due to the stimulating and stimulating effect on biomarkers. Nevertheless, cannabis seems to effectively relieve “internal pressure”.
The active ingredients contained are primarily the psychoactive cannabinoids (such as THC = dronabinol) and the non-psychoactive cannabidiol (CBD).
2.1. Cannabinoids (THC)
2.1.1. Cannabinoids affect dopamine, serotonin, and acetylcholine in the PFC
Cannabinoids affect dopamine, serotonin, and acetylcholine in the PFC.16 Deficits in working memory, attentional function, and reversal learning with THC use as a drug appear to be mediated by activation of the CB1 cannabinoid receptor.
The positive effect of THC as a medicine or in microdosing use (without intoxication) could possibly be mediated via this. More detailed information on this is lacking.
2.1.2. Increase in the level of cortisol
One connection could be that THC (in non-THC-dependent individuals) increases cortisol levels in a dose-dependent manner.1718 This had previously been found in animal studies.19 Another study, however, found no cortisol change due to THC.20
Cannabinoids generally influence the activation of neurotransmitters such as GABA and glutamate, which regulate the stress axis; THC activates the HPA axis.21 When cortisol level was measured 4 hours after THC administration, it was not (anymore) elevated.22
In drug users who were abstinent for at least 2 weeks at the time of the study, no cortisol increase to THC was detected.23 This could possibly be a consequence of receptor downregulation.
2.1.3. HPA axis shutdown by THC-mediated cortisol increase?
We consider it theoretically conceivable that (infrequent and mild) THC intoxication could possibly produce a relaxing effect in that the resulting increased cortisol release effectively shuts down the HPA axis in ADHD-HI and ADHD-C sufferers. ADHD-HI and ADHD-C sufferers (with hyperactivity) have a flattened cortisol response to stress, whereas ADHD-I sufferers show excessive cortisol release from acute stress. Since cortisol not only mediates stress symptoms, but (as the last stress hormone of the HPA axis to be released in time) also has the task of down-regulating the HPA axis again, (at least some of) ADHD-HI and ADHD-C sufferers lack the regular deactivation of the HPA axis by cortisol.
However, this is presumably true only in occasional users, not in continuous users of THC, in whom receptor downregulation may prevent this effect.
This hypothesis correlates with THC being primarily consumed by ADHD-HI sufferers. ADHD-I sufferers, who do not have an HPA axis shutdown problem due to their excessive cortisol response to stress, are less likely to consume THC.
This hypothesis further correlates with reports from ADHD-HI sufferers that after THC use, the resulting calm lasts for about 2 to 3 days.
2.1.4. Decreased cortisol response to acute stressors
In continuous users who consume THC every day or almost every day, (probably due to downregulation or upregulation *⇒ Downregulation / Upregulation the cortisol response to acute stressors is reduced.
In long-term users, THC can therefore no longer trigger downregulation of the HPA axis by cortisol in the absence of an increase in cortisol release.
The decreased cortisol response to acute stress in THC continuous users2425 could explain the assumption of a high cortisol output by THC and a consequent special THC affinity of people with flattened cortisol response to stress.
Schizophrenia-affected individuals who had previously used cannabis showed lower morning cortisol elevation on awakening (CAR) than schizophrenia-affected individuals without previous cannabis use, whose levels were comparable to those of nonaffected individuals.26 This finding may also reflect the affinity of people with flattened cortisol responses to stress.
Independently of this, THC has an anxiolytic (anxiety-reducing) effect up to certain doses. At higher doses, however, it has an anxiety-promoting effect.27
Anti-anxiety drugs lead to less intense activation of the HPA axis by means of calming the amygdala.
2.1.5. Reduction of the amygdala volume
Evidence suggests that prolonged cannabis use decreases amygdala and hippocampal volume.28
It would be conceivable that reducing the size of the amygdala would simultaneously reduce its activity and that such a reduction in amygdala activity would reduce the intensity of stress responses.
Unfortunately, this is not the only effect of cannabis.
Problematically, even infrequent use can lead to dependence, which is why therapy cannot be advocated by infrequent cannabis use.
On the other hand, ADHD-HI sufferers (especially those with comorbid aggression disorders) report that even continuous use gives them an inner serenity that cannot be achieved otherwise.
Continuous cannabis users with ADHD are conspicuous for increased hospitalization and decreased use of medication and behavioral therapy.29
2.1.6. Cannabinoids change the EEG: Alpha increased, beta decreased
Cannabis increases activity in the alpha band of the brain and decreases activity in the beta band, as well as the other frequencies of the EEG.303132
The fact that cannabis increases alpha activity (which is required for concentration) beyond acute consumption while simultaneously reducing the activity of all other frequencies (i.e., also beta frequencies which, when increased, prevent adequate relaxation, e.g., thought-circling) could explain why cannabis is perceived as pleasant by many ADHD sufferers. That this effect is not beneficial in the long run with respect to the overall constitution is understandable. That this effect has a short-term relieving effect for ADHD sufferers is also plausible.
2.1.7. Cannabis abuse correlates with ADHD symptoms, not cognitive problems
One larger study found a correlation between cannabis abuse and ADHD symptoms such as hyperactivity/impulsivity and inattention. Whether cannabis abuse was the cause/trigger or consequence of the symptoms (self-medication) was not investigated. Further, the study found no correlation between cannabis abuse and cognitive problems.33
2.2. Cannabidiol (CBD)
Cannabidiol (CBD) is not psychoactive and is another relevant constituent of cannabis along with cannabinoids (THC).
Cannabidiol reduces anxiety after a stress test, but not before or during.34
A comparison of different doses showed that cannabidiol only reduced anxiety at low doses (300 mg) during and after a stress test, but not at 600 mg or 900 mg.35 Another study comparing the effect of 600 mg cannabidiol against placebo in social phobia sufferers found an anxiety-reducing effect during the stress test, although still more anxiety was reported than in non-sufferers.36
Anxiety sufferers told us that taking CBD hemp oil allowed them to significantly reduce their anxiolytic medications.
An effect of CBD in ADHD has not been reported.
2.3. THC drugs for ADHD
THC medications (which do not cause intoxication due to dosage, but rather are designed to have a consistent lasting effect) can have a positive effect on ADHD symptoms.
- THC is a preferred drug of ADHD sufferers. The relaxing effect is appreciated, which some sufferers describe as lasting for 2 to 3 days after use. This long-lasting effect could possibly be the result of a down-regulation of the HPA axis by the THC-induced cortisol release (in occasional users, not in continuous users).
- THC stimulates the reward / reinforcement center of the brain. This is one of the central neurological problems in ADHD: a too low dopamine level in the striatum.
- Whether THC actually increases dopamine output in the brain’s reward center, the striatum, is unclear. Some studies affirm this,3738 others deny it.39 Summing up Pertwee.40 Since dopaminergic intoxicating drugs function by increasing dopamine levels in the brain’s reinforcement center, and intoxication is based on a massive short-term excess of dopamine, an increase in dopamine levels by THC would be plausible. Apart from the fact that when consumed as a drug, there is a rapid increase in dopamine above the functional level, whereas drugs cause a slow increase to a constant (functional) level, the short-term effect also changes at the latest with continuous use: in THC-dependent individuals, dopamine levels in the striatum are reduced.41 This opens the question whether this is a consequence of addiction (which would be conclusive, since a permanently elevated dopamine level causes a downregulation of dopamine transporters, resulting in an increased presynaptic reuptake of dopamine, making less dopamine available in the synaptic cleft to the postsynapse), or is a consequence of an excess of dopamine level in the PFC, since this results in a decreased dopamine level in the striatum. A persistently too low level of dopamine in the striatum is likely to cause an increase in DAT as an upregulation counterreaction. It is further conceivable that the decreased dopamine level in the striatum is the cause that once made THC attractive to addicts. Heinz answers this for alcohol addicts in the direction of both and, although the relationships are infinitely more complex.42
However, it should be noted that cannabinoids are dopaminergic drugs and ADHD sufferers also have a far above-average affinity with regard to nicotine and alcohol, which also have a dopaminergic effect.
- THC is established as a medication for Tourette syndrome.43 Tourette’s is a very common comorbidity of ADHD.
- Individuals report that “regular” low THC use significantly reduces their risk of migraine. Migraine is a common comorbidity of ADHD.
We know of several patients who were able to largely solve severe ADHD symptoms with strong comorbid disorders (including bipolar) with cannabinoid drugs (medical hash) - at least better than with stimulants and other drugs. (Private prescription, not reimbursed by the health insurance; 130 € / month; as of fall 2017).
Meanwhile, more and more physicians are aware of the medical benefits of prescribing THC for ADHD sufferers for whom other medications have not been successful and are willing to prescribe it.
2.4. ADHD symptom improvement through microdosing (microdosing)
Against the above thesis of HPA axis shutdown by THC-mediated massive stress-induced cortisol increase (at least as the sole mode of action) speaks that affected persons (and indeed also those of the ADHD-I subtype) report a significant symptom improvement by microdosing THC or cannabidiol.
Microdosing in this context means a dosage so low that (cumulatively)
- No noise sensation occurs at all
- The short-term memory is not impaired
- No performance impairment / flabbiness occurs.
Several affected individuals report that a microdose of cannabinoids / cannabidiol - usually using a vaporizer - significantly increases their focus. If short-term memory impairment occurs, it is reported that an even lower dosage can improve this.
It is further reported that the optimal mixing ratio between cannabinoids and cannabidiol varies from individual to individual and the degree of symptom improvement is influenced by the individual adjustment of the mixing ratio. A very low cannabidiol content has been reported to be beneficial.
These notes are not a recommendation for self-medication!
According to §§ 29 ff. BtMG, unauthorized cultivation, production, trade, import, export, distribution, sale, other placing on the market, acquisition and possession of all parts of the cannabis plant is punishable in Germany.
The basic principle of microdosing can be scientifically considered as the difference between drug (high dosage leads to intoxication) and medication (low dosage remedies deficit). This principle is exactly what is known to make the difference of amphetamine as a drug or as a medication (e.g. in ADHD).
However, amphetamine medications or amphetamine-related medications, as widely used in ADHD, are subject to multiple reviews and very precise dosing due to medical approval procedures. This is not possible with natural substances, so that there is always a risk of incorrect dosing.
Therefore, self-medication - also independent of the legal issues that arise - is generally not advisable.
In addition, it should be noted that the alleged medical use sometimes serves merely as a sham to mask drug abuse.
If the usual medications for ADHD (amphetamine, methylphenidate, guanfacine, atomoxetine, bupropion, nicotine) do not help sufficiently, medical cannabis can be prescribed by a doctor. It is known that in some cases this is the most effective remedy for ADHD.
Medical cannabis has a defined active ingredient content (which is measured and documented batch by batch) and can therefore be precisely dosed.
For this application area, the explanation of microdosing is instructive.
3. Amphetamines / Mesamphetamines
Amphetamine drugs can significantly lower stress levels. However, abuse of amphetamine drugs leads to a long-term increased stress response.44
Meprobromate, a sedative from the 1960s (now illegal because of its addictive potential), reduces the adrenaline surge under stress.45
“Crystal meth” is a mesamphetamine and is very related to amphetamine’s mode of action as an ADHD medication.46
It is reported that the use of illicit substances such as cocaine and amphetamines can have a positive effect on concentration deficits (“self-medication”).47
It goes without saying that this cannot result in legal justification. However, it will have to be taken into account in the context of the question of guilt (severity of guilt), which cannot prevent a penalty, but can possibly mitigate it.
Cocaine inhibits the reuptake transporters of dopamine, norepinephrine, and serotonin.
- Alcohol increases dopamine levels
- Alcohol significantly increases epinephrine levels for at least 12 hours after ingestion of moderate amounts of alcohol (1.43 g/kg body weight). Norepinephrine levels are significantly elevated, especially in the first 6 hours, and tend to be elevated thereafter.49
- Excessive alcohol consumption can increase epinephrine and norepinephrine levels for up to one week. Similarly, increased levels of 17-hydroxycorticosteroid were found. 0.5 g of chloromethiazole decreased catecholamine levels.50
Microdoses of LSD (100 nanograms/kg and less) appear to raise serotonin levels, while even slightly higher doses cause a decrease in serotonin.51
An online survey revealed a significantly higher subjective effect size of microdosing with psychedelic substances such as LSD than a treatment considered conventional for psychological and physiological diagnoses, especially in the case of52
- Anxiety disorders
- Autism spectrum disorders (ASD) (though not significant here)
- Personality disorders (though not significant here).
In contrast, higher, regularly psychedelic doses were judged to be better than microdosing at
No significant difference was found for
- Physiological disorders.
Another study reported fewer side effects under microdosing of LSD (10 micrograms of LSD every third day was tested in more than 1000 subjects with a broad spectrum of disorders over 18 months) compared to conventional treatment.53
Silva, Szobot, Shih, Hoexter, Anselmi, Pechansky, Bressan, Rohde (2014): Searching for a neurobiological basis for self-medication theory in ADHD comorbid with substance use disorders: an in vivo study of dopamine transporters using (99m)Tc-TRODAT-1 SPECT. Clin Nucl Med. 2014 Feb;39(2):e129-34. doi: 10.1097/RLU.0b013e31829f9119. ↥
Pomerleau (1997): Co-factors for smoking and evolutionary psychobiology. Addiction, 92, 397– 408, zitiert nach Barkley, DuPaul, McMurray (1991): Attention deficit disorder with and without hyperactivity: Clinical response to three dose levels of methylphenidate. Pediatrics, 87, 519–531, zitiert nach Diamond: Attention-deficit disorder (attention-deficit/hyperactivity disorder without hyperactivity): A neurobiologically and behaviorally distinct disorder from attention-deficit (with hyperactivity), Development and Psychopathology 17 (2005), 807–825,, Seite 811, Seite 811 ↥
Diamond: Attention-deficit disorder (attention-deficit/hyperactivity disorder without hyperactivity): A neurobiologically and behaviorally distinct disorder from attention-deficit (with hyperactivity), Development and Psychopathology 17 (2005), 807–825, Seite 811 mit weiteren Nachweisen ↥
Frankenhaeuser, M. (1971). Behavior and circulating catecholamines. Brain Research, 31(2), 241-262. http://dx.doi.org/10.1016/0006-8993(71)90180-6, Seite 255, mit weiteren Nachweisen ↥
Ashare, Hawk (2012): Effects of smoking abstinence on impulsive behavior among smokers high and low in ADHD-like symptoms; Psychopharmacology (Berl). 2012 Jan; 219(2): 537–547. doi: 10.1007/s00213-011-2324-2, PMCID: PMC3184469, NIHMSID: NIHMS300102, PMID: 21559802, n = 56 ↥
Brandt, Rehm, Lev-Ran (2018): Clinical Correlates of Cannabis Use Among Individuals With Attention Deficit Hyperactivity Disorder. J Nerv Ment Dis. 2018 Sep;206(9):726-732. doi: 10.1097/NMD.0000000000000877. ↥
Ranganathan, Braley, Pittman, Cooper, Perry, Krystal, D’Souza (2008): The effects of cannabinoids on serum cortisol and prolactin in humans; Psychopharmacology (Berl). 2009 May;203(4):737-44. doi: 10.1007/s00213-008-1422-2. n = 76 ↥
Childs, Lutz, de Wit (2017): Dose-related effects of delta-9-THC on emotional responses to acute psychosocial stress. Drug Alcohol Depend. 2017 Aug 1;177:136-144. doi: 10.1016/j.drugalcdep.2017.03.030. n = 42 ↥
Hollister, Moore, Kanter, Noble (1970): 1 -tetrahydrocannabinol, synhexyl and marijuana extract administered orally in man: catecholamine excretion, plasma cortisol levels and platelet serotonin content. Psychopharmacologia. 1970;17(4):354-60. ↥
Cuttler, Spradlin, Nusbaum, Whitney, Hinson, McLaughlin (2017): Blunted stress reactivity in chronic cannabis users. (Psychopharmacology (Berl). 2017 Aug;234(15):2299-2309. doi: 10.1007/s00213-017-4648-z. ↥
van Leeuwen, Creemers, Greaves-Lord, Verhulst, Ormel, Huizink (2011): Hypothalamic-pituitary-adrenal axis reactivity to social stress and adolescent cannabis use: the TRAILS study. Addiction. 2011 Aug;106(8):1484-92. doi: 10.1111/j.1360-0443.2011.03448.x. ↥
Monteleone, Di Filippo, Fabrazzo, Milano, Martiadis, Corrivetti, Monteleone, Maj (2014): Flattened cortisol awakening response in chronic patients with schizophrenia onset after cannabis exposure. Psychiatry Res. 2014 Feb 28;215(2):263-7. doi: 10.1016/j.psychres.2013.12.016. ↥
Childs, Lutz, de Wit (2017): Dose-related effects of delta-9-THC on emotional responses to acute psychosocial stress. Drug Alcohol Depend. 2017 Aug 1;177:136-144. doi: 10.1016/j.drugalcdep.2017.03.030, n = 42 ↥
Yücel, Solowij, Respondek, Whittle, Fornito, Pantelis, Lubman (2008): Regional brain abnormalities associated with long-term heavy cannabis use. Arch Gen Psychiatry. 2008 Jun;65(6):694-701. doi: 10.1001/archpsyc.65.6.694. ↥
Patel, Patel, Shah, Kaur, Mansuri, Makani (2018): Is Cannabis Use Associated With the Worst Inpatient Outcomes in Attention Deficit Hyperactivity Disorder Adolescents? Cureus. 2018 Jan 7;10(1):e2033. doi: 10.7759/cureus.2033. n = 11232 ↥
Fink (1976): Effects of acute and chronic inhalation of hashish, marijuana, and delta 9-tetrahydrocannabinol on brain electrical activity in man: evidence for tissue tolerance. Ann N Y Acad Sci. 1976;282:387-98. ↥
Struve, Straumanis, Patrick, Price (1989): Topographic mapping of quantitative EEG variables in chronic heavy marihuana users: empirical findings with psychiatric patients. Clin Electroencephalogr. 1989 Jan;20(1):6-23 ↥
Petker, DeJesus, Lee, Gillard, Owens, Balodis, Amlung, George, Oshri, Hall, Schmidt, MacKillop (2020): Cannabis use, cognitive performance, and symptoms of attention deficit/hyperactivity disorder in community adults. Exp Clin Psychopharmacol. 2020 Feb 27. doi: 10.1037/pha0000354. PMID: 32105137. n = 1.008 ↥
Zuardi, Rodrigues, Silva, Bernardo, Hallak, Guimarães, Crippa (2017): Inverted U-Shaped Dose-Response Curve of the Anxiolytic Effect of Cannabidiol during Public Speaking in Real Life. Front Pharmacol. 2017 May 11;8:259. doi: 10.3389/fphar.2017.00259. eCollection 2017. ↥
Bergamaschi, Queiroz, Chagas, de Oliveira, De Martinis, Kapczinski, Quevedo, Roesler, Schröder, Nardi, Martín-Santos, Hallak, Zuardi, Crippa (2011): Cannabidiol reduces the anxiety induced by simulated public speaking in treatment-naïve social phobia patients. Neuropsychopharmacology. 2011 May;36(6):1219-26. doi: 10.1038/npp.2011.6. n = 24 ↥
Bossong, van Berckel, Boellaard, Zuurman, Schuit, Windhorst, van Gerven, Ramsey, Lammertsma, Kahn (2009): Delta 9-tetrahydrocannabinol induces dopamine release in the human striatum.; Neuropsychopharmacology. 2009 Feb;34(3):759-66. doi: 10.1038/npp.2008.138. ↥
Stokes, Mehta, Curran, Breen, Grasby (2009): Can recreational doses of THC produce significant dopamine release in the human striatum? Neuroimage. 2009 Oct 15;48(1):186-90. doi: 10.1016/j.neuroimage.2009.06.029. ↥
van de Giessen, Weinstein, Cassidy, Haney, Dong, Ghazzaoui, Ojeil, Kegeles, Xu, Vadhan, Volkow, Slifstein, Abi-Dargham (2016): Deficits in striatal dopamine release in cannabis dependence. Mol Psychiatry. 2016 Mar 22. doi: 10.1038/mp.2016.21. Niedrige Probandenzahl von n = 21 ↥
Parrott, Montgomery, Wetherell, Downey, Stough, Scholey (2014): MDMA, cortisol, and heightened stress in recreational ecstasy users. Behav Pharmacol. 2014 Sep;25(5-6):458-72. doi: 10.1097/FBP.0000000000000060. ↥
Sören Schmidt und Franz Petermann, ADHS über die Lebensspanne – Symptome und neue diagnostische Ansätze, Zeitschrift für Psychiatrie, Psychologie und Psychotherapie, 59 (3), 2011, 227–238, Seite 233 ↥
Hutten, Mason, Dolder, Kuypers (2019): Self-Rated Effectiveness of Microdosing With Psychedelics for Mental and Physical Health Problems Among Microdosers. Front Psychiatry. 2019 Sep 13;10:672. doi: 10.3389/fpsyt.2019.00672. eCollection 2019. n = 3590 ↥