II - Classification

Note: Content is not peer-reviewed, does not constitute medical advice, and is provided for educational/interest purposes only.

8. Neurodevelopmental disorders

MTA study (ADHD)

• MTA study initial results vs long-term. MTA original. 

• At 3 years meds or combo are no longer better than the rest. Link. DOI: 10.1097/CHI.0b013e3180686d48

• Delinquent behavior and emerging… - At 3 years more delinquency in med group. Link. DOI: 10.1097/chi.0b013e3180686d96

• The MTA at 8 years… At 8 years meds are no better. Link. DOI: 10.1097/CHI.0b013e31819c23d0

• 2025 - “After 14 months of treatment, the children who took Ritalin every day had significantly fewer symptoms… by 36 months, that advantage had faded completely, and children in every group, including the comparison group, displayed exactly the same level of symptoms.” Link. (NYT, "Have We Been Thinking About A.D.H.D. All Wrong?")

• 2025 - Swanson MTA author “If you’re honest, you should tell kids that, look, if you’re interested in next week or next month or even the next year, this is the right treatment for you. But in the long run, you’re going to be shorter. How many kids would agree to take medication? Probably none.” Link. (NYT, "Have We Been Thinking About A.D.H.D. All Wrong?")

• 2025 - Swanson MTA author argues that individuals using stimulants quickly develop tolerance, causing their effects to dissipate until they are no longer present. He suggests a single morning dose of a short-acting stimulant to prevent this. Link.

ADHD

• Stimulants don’t improve academic performance. Link. DOI: 10.1037/ccp0000725

• ADHD diagnosis on its own leads to worse outcomes. 393 ADHD adolescents matched with 393 others with similar ADHD symptoms, the ones with the diagnosis do worst. Link. DOI: 10.1001/jamanetworkopen.2022.36364

• 2024 - 23% of boys have received a diagnosis with ADHD by age 17 (number not included in the paper but calculated from Table 1). Link. DOI: 10.1080/15374416.2024.2335625

• 2025 - 93.7% adults with ADHD get treatment with meds - Link. DOI: 10.1176/appi.ajp.20240492

• Review of all treatment (meds, not meds) for ADHD. Only meds have significant evidence at 12 weeks. CBT and Neurofeedback are pretty good at medium term. Few papers and high variability. Link. DOI: 10.1016/s2215-0366(24)00360-2

• 2025 - "38% of Stanford undergraduates are registered as having a disability" mostly driven by ADHD Link (Atlantic "Accommodation Nation")

• 2025 - Methylphenidate doesn’t improve exam scores. Link. DOI: 10.1007/s00213-025-06864-1

• 2025 - Data on stimulants use in adult and adolescents in the US 2005 to 2024 (figure 2). Link (U.S. Population Prevalence ofPrescription Psychiatric Medication Use AmongChildren and Adolescents, 2006-2023)

  • Adults - 1.1% in 2005, 3.1% in 2024, 181.82% increase

  • Teens - 5.0% in 2005, 5.3% in 2024

Risk of stimulants

• 29% of kids on stimulants have insomnia (vs 10% in untreated) - Link. DOI: 10.1089/cap.1999.9.157

• Longitudinal study (1998) of 492 children over 20 years. ADHD with and without stimulants and controls. Examined outcomes in terms of drug dependence in adulthood. Stimulant use associated with: Increased smoking dependence by 53%, Increased cocaine dependence by 50%. Post. Link. DOI: 10.1177/002221949803100603

• 2024 - Amphetamines x5 psychosis and mania. Link. DOI: 10.1176/appi.ajp.20230329

• 2025 - 25.3% of those prescribed stimulants reported misuse - Link. DOI: 10.1001/jamapsychiatry.2025.0054

• 2025 - RR of stimulants for all cause mortality: 3.53. Link. DOI: 10.1097/JCP.0000000000002066

ADHD biology

• 2025 - We have a clinical definition of A.D.H.D. that is increasingly unanchored from what we’re finding in our science.” Link. (NYT, "Have We Been Thinking About A.D.H.D. All Wrong?")

• 2025 - “the evidence for each of these A.D.H.D. biomarkers has faltered.” Link. (NYT, "Have We Been Thinking About A.D.H.D. All Wrong?")

9. Neurocognitive disorders

Delirium

• 2025 - Dexmedetomidine + melatonin, best to prevent delirium. Risperidone was #2, possibly against the criticism of antipsychotics in delirium. Link. DOI: 10.1111/anae.16757

• Dexmedetomidine (Precedex) - Dexmedetomidine acts by binding to alpha2-adrenergic receptors in the brainstem, which reduces the release of norepinephrine. It's primarily known for its sedative properties. It is used for delirium. Similar to clonidine.

2025 - APA guidelines for delirium. Link. DOI: 10.1176/appi.books.9780890428023

• APA suggests that dexmedetomidine be used rather than other sedating agents to prevent delirium in patients who are undergoing major surgery or receiving mechanical ventilation in a critical care setting.

• APA suggests that melatonin and ramelteon not be used to prevent or treat delirium.

• APA says that antipsychotics shouldn’t be used to reduce delirium but to treat acute agitation that didn’t respond to verbal de-escalation

Dementia

• 2025 - Decreasing rate of dementia. "The age-specific dementia prevalence rates declined by approximately two-thirds in the US during 1984 to 2024". Link. DOI: 10.1001/jama.2025.1897

Beers criteria

• 2025 - Alternative Treatments to Selected Medications in the 2023 American Geriatrics Society Beers Criteria. "Instead of replacing a potentially inappropriate medication with a "better" one, consider non-pharmacologic strategies." For insomnia: avoid Benzos, Z-drugs, antihistamines, TCA, consider CBTi. For anxiety avoid benzodiazepines, antihistamines, TCAs, barbiturates, consider tapering benzos, self-help books, SSRIs, SNRIs. Delirium and agitation: avoid antipsychotic. Link. DOI: 10.1111/jgs.19500

10. Substance-related disorders

Mechanism of action (clarification)

• Alcohol mostly affects GABA-A (but also GABA-B). BZD, Barb, Z-drugs are mostly GABA-A. Baclofen and GHB are mostly GABA-B. 

• Alcohol leads to the release of endorphins from the hypothalamus and pituitary. Endorphins activate opioid receptors. Naltrexone blocks this activation.

• Gabapentin reduces Glutamate more than activate GABA-A or GABA-B (by being Ca Channel Blocker)

• Campral is an NMDA antagonist (by being a Ca Channel Blocker)

Risk of stimulants

• 2024 - Amphetamines x5 psychosis and mania. Link. DOI: 10.1176/appi.ajp.20230329

Risk of sedatives

• 2024 - Some risk of addiction with Pregabalin. Link. (BBC, "Rising deaths linked to pregabalin misuse sparks review")

• 2024 - 76,137 Benzodiazepine patients vs. matched cohort found to have increased risk of mood and substance use disorders "Our findings demonstrate a concerning association between BZD use and an increased risk of being diagnosed with various mood disorders and SUDs." Link. DOI: 10.1016/j.dadr.2024.100270

• 2025 - Benzodiazepines have seen a significant decrease in use in favor for other sedatives (gabapentin, trazodone, hydroxyzine, doxepin, pregabalin). Post. CMS Data. 

• 2025 - Benzodiazepines prescription linked to increase future substance misuse. "[Benzodiazepines] initiation was associated with a small but robust increase in absolute and relative risks of developing alcohol- and drug-related problems." Link. DOI: 10.1177/02698811251373069

Other

• 2024 - Contrary to common belief, vaping doesn't appear to have replaced smoking, but added to it - Link. DOI: 10.1016/S2468-2667(24)00183-X External Link

Antidepressants for SUD

• 265 patient w/ alcohol use dz randomized to placebo v citalopram + therapy. Citalopram did not lead to higher abstinence, but was associated with more heavy drinking days, an average of 13.02 versus 7.78 for placebo. Link. DOI: 10.1111/acer.12802

• 229 patients w/ meth use dz randomized to a placebo, sertraline, contingency management or combinations found that “participants in the sertraline-only condition provided significantly more metabolite-positive urine samples.” Link. DOI: 10.1016/j.drugalcdep.2006.03.005

Self-medication model

• A 2020 review noted there is "a lack of rigorous empirical evidence is support of the self-medication model." Link. DOI: 10.1037/adb0000552

• 1985 - Origin of self-medication hypothesis by Dr. Khanztian. It wasn't untrue but it is interesting nonetheless. Post. Link. DOI: 10.1176/ajp.142.11.1259

GLP-1 for alcohol use disorder

• 2025 - Review. GLP-1s "show beneficial effects on consumption, relapse, and alcohol-related morbidity", did better than naltrexone. Link. DOI: 10.1016/j.eclinm.2025.103645 

11. Schizophrenia and others thought disorders

Preventing illness

• 2025 - "No evidence that any of the investigated active interventions had a sustained and robust effect on any of the investigated outcomes, compared to control interventions." Link. DOI: 10.1038/s41380-025-02902-8

Antipsychotics

• Self-reported reasons for antipsychotic use, NHANES 2013-2018. Schizophrenia 9.6%. Bipolar 25.5%. Depression 32.3%. Insomnia 16.6%. Unspecified (psychosis/mood/anxiety) 22.2%. Aka: 90+% of antipsychotics not for schizophrenia. Link. DOI: 10.1186/s12888-020-02895-4

• SGAs cause similar tardive dyskinesia as FGAs in long term. N=28,051 followed over many years. Link. DOI: 10.1097/YCO.0b013e3282f53132

• Only 6.3% of #1 antipsychotic quetiapine is for psychosis. Post. Link. DOI: 10.1136/bmjopen-2015-010861

• 2025 - Clozapine no better than other SGAs in this system review. The review "did not provide evidence for superior efficacy of clozapine." Link. DOI: 10.1016/S2215-0366(25)00001-X 

• 2025 - Cobenfy failed as an adjunct treatment to SGAs in ARISE trial. Link.

• 2025 - Nuplazid (Pimavanserin) for negative symptoms of schizophrenia failed trial - ADVANCE-2 trial - Link. DOI: 10.1093/schbul/sbaf034

• 2025 - 10 year long study of individuals with schizophrenia continuously on LAI versus others. LAI did better and seemingly did better as time went on (in contrast to Wunderick) - Link. DOI: 10.1176/appi.ajp.20240321 

• 2025 - Data on antipsychotics use in adult and adolescents in the US 2005 to 2024 (figure 3). Link. (U.S. Population Prevalence of Prescription Psychiatric Medication Use Among Children and Adolescents, 2006-2023)

  • Adults - 1.9% in 2005, 3% in 2024, 57.89% increase

  • Teens - 1.3% in 2005, 1.1% in 2024

• 2025 - HAMLETT: dose reduction/discontinuation of antipsychotic study. What is striking to me in that trial is how most in the discontinuation/reduction still end up on similar amount of meds in the long-term than the continuation group. The paper highlighted some potential long-term advantage of that initial discontinuation/reducation. Link. DOI: 10.1001/jamapsychiatry.2025.2525

Diagnostic issues

• 2023 - Risk of getting a schizophrenia spectrum disorder after ED visit for substance use (without psychosis). Hazard ratios: amphetamine 28.4, polysubstance 18.7, Cannabis 14.3, Cocaine 13.2, Alcohol, 6.1, Other 13.3 (obviously the HR are much higher when the ED presentation included psychosis). N= 9.8M. Link. Post. DOI: 10.1001/jamapsychiatry.2023.3582 

• 2025 - CBD & psychosis. THC +/- CBD (CBD, which is sometimes thought to maybe help psychosis/cognition/mental illness), individuals randomized with CBD had worse cognition (verbal recall) and more psychotic symptoms. Link. DOI: 10.1038/s41386-025-02175-3

• 2025 - Study of autoimmune psychosis notes, n= 164. 39% were initially misdiagnosed as primary psychosis, severe cognitive dysfunction was present in 75%, 64% were also diagnosed with catatonia. Link. DOI: 10.1016/j.schres.2025.04.024

12a. Bipolar and related disorders

Treatment

• 2025 - Swedish cohort, N=105,495. "Only lithium was associated with lower depression-related hospitalisation risk." Li was "superior" to antidepressants, antipsychotics, lamotrigine, quetiapine. Antidepressant-only treatment "was associated with increased risk of mania-related hospitalisation." Link. DOI: 10.1016/S2215-0366(24)00411-5 

• Data on mood stabilizers use in adult and adolescents in the US 2005 to 2024. Link. (U.S. Population Prevalence ofPrescription Psychiatric Medication Use AmongChildren and Adolescents, 2006-2023)

  • Adults - 1.7% in 2005, 2.6% in 2024, 52.94% increase

  • Teens - 1.0% in 2005, 0.8% in 202

  • This doesn’t take into account indication which is important to note for mood stabilizers

Diagnostic

• 2025 - Pediatric bipolar down ~90% from 2013 to 2021 - Post. Link. DOI: 10.1016/j.jaac.2024.08.008

13. Anxiety disorders

• Lifetime prevalence of 3x more prevalent in high income countries (5.0 v 1.6). US was #3 at 7.8%. Post. Link. DOI: 10.1001/jamapsychiatry.2017.0056 

• Randomized trial: all participants receive Lexapro for social anxiety. Half are told it's an active placebo, half are told it's a medication. Results show a 3.6 times higher response rate in those told it's a medication. Conclusion "effects of escitalopram were markedly influenced by verbal suggestions [that] favors a biopsychosocial over a biomedical explanatory model for SSRI efficacy". Post. Link. DOI: 10.1016/j.ebiom.2017.09.031 

• 2025 - RCT of panic disorder (though most also had depression). N=60. Sertraline + placebo v Sertraline + propranolol. Conclusion: "using propranolol alongside sertraline reduces the severity of panic attacks." Link. DOI: 10.1186/s40780-025-00466-x

14. Obsessive-compulsive disorders

15. Trauma and stress disorders

• Children in households with one natural and one stepparent are more likely to be victim of abuse“We are not talking about a small effect: preschoolers in stepparent-natural-parent homes, for example, are estimated to be 40 times as likely to become abuse statistics as like-aged children living with two natural parents (i.e., about 40 chances in 3000 vs. 1 in 3000; Figure I).” Link. DOI: 10.1016/0162-3095(85)90012-3 

• 2025 - Comparing checklists to clinical interviews, only 21% of reported trauma was true positive, (79% was false positive). For symptoms, only 34% were true positive, (65.9% were false positive). A powerful argument against checklist for PTSD. Link. DOI: 10/1089/cap.2024.0126.

16. Dissociative disorders

17. Somatic symptom disorders

• 2025 - Review/meta-analysis non-surgical treatments for low back pain (LBP). Acute LBP (<12 weeks): NSAIDs were found to be efficacious. Exercise, steroid injections, and Tylenol were not efficacious. Chronic LBP (≥12 weeks): Exercise, chiropractor, taping, capsaicin (TRPV1 agonists), and antidepressants were efficacious. Antibiotics and anaesthetics were not efficacious. Inconclusive Evidence: Evidence for the majority of other treatments (e.g., muscle relaxants, opioids, and acupuncture) is inconclusive due to low certainty or small samples. Link. DOI: 10.1136/bmjebm-2024-112974

18. Factitious disorders and malingering

• Up to 81% of individuals intentionally withhold “medically relevant information” from providers (not malingering but relevant). Link. DOI: 10.1001/jamanetworkopen.2018.5293

• Update the psychogenic seizure analogy -> 25%-50% evaluated for status epilepticus are ultimately diagnosed with functional seizures. 30%–40% of children with functional seizures also have epilepsy. Link. DOI: 10.1212/WNL.0000000000214466

19. Eating and elimination disorders

• 2023 - Review/meta-analysis pharmacotherapy for bulimia. Mentions only fluoxetine is approved and is considered first-line. No pharmacotherapy was significant at improving remission rates. Link. DOI: 10.1016/j.psychres.2023.115357

20. Sleep disorders

21. Sexual and gender disorders

22. Disruptive and impulse control disorders

23. Personality disorders

• 2024 - Review/meta-analysis. For individuals with BPD, psychotropics medications were not shown to be effective whether co-pathology is present or not (low certainty evidence). Link. DOI: 10.1192/bjp.2024.172

24. Treatment Principles

•  Placebo

  • TMS v Sham TMS. VA study, n=164. No significant difference, in part due to robust sham TMS response. Remission rate TMS 40.7%, sham 37.4%. "placebo response has been increasing over time". Link. DOI: 10.1001/jamapsychiatry.2018.1483

  • 2024 - Systematic review & meta-analysis of placebo comparing effect size of placebo in different disorders. MDD and GAD had highest effect size. Link. DOI: 10.1001/jamapsychiatry.2024.0994

25. Psychotherapies

• Will likely add a section on Behavioral therapy interventions

• Will likely add a subsection on concerns about excessive validation. Link 1, Link 2.

• 2025 - Psychodynamic "non-inferior to CBT in reducing depressive symptoms". N=290. Link. DOI: 10.1016/j.jad.2024.10.122

• 2024 - N-1649 VA study showing that what is considered "adequate" therapy is often too short to be effective. Link. DOI: 10.1037/ser0000830 

• Consumer Report survey of mental health in 7,000 subscribers. Link. DOI: 10.1037/0003-066X.50.12.965

  • 2,900 saw mental health professionals. 426 felt “very poor” prior to treatment, 87% of those improved. Among the 786 who felt “fairly poor”, 92% improved. The longer the treatment the more improvement subscribers had, measured up to more than two years.

  • “There was no difference between psychotherapy alone and psychotherapy plus medication for any disorder.” 

  • “psychologists, psychiatrists, and social workers did equally well”

  • “Alcoholics Anonymous (AA) did especially well… significantly bettering mental health professionals.”

  • “all forms of psychotherapies do about equally well”