Supply & Generics

Amphetamine mixed salts shortage — active, entering its fourth year

psychopharmref.com · 2026-07-02

Generic amphetamine mixed salts, in both immediate-release and extended-release formulations, remain listed as actively short by the FDA and the American Society of Health-System Pharmacists (ASHP) as of June 2026 (drugs.com, date unavailable; ASHP, date unavailable). The shortage, which began in 2022, has persisted across multiple manufacturers and has not resolved despite modest improvements in supply reported earlier this year. Lisdexamfetamine dimesylate capsules and tablets are separately listed as short in the same databases, meaning two of the most commonly prescribed first-line agents for ADHD are simultaneously constrained (drugs.com, date unavailable).

ASHP's Q1 2026 figures place the total number of active drug shortages at 223, with approximately 15 percent involving controlled substances — a category that captures the stimulant class disproportionately (ASHP, date unavailable). That figure represents an upward trend for the second consecutive quarter, though ASHP notes it remains below the 2024 peak (Becker's Hospital Review, 2026). The FDA Drug Shortages database and ASHP track manufacturer-level reporting, but specific manufacturer names and attributed root causes — manufacturing delays, DEA (the US Drug Enforcement Administration, which sets annual production quotas for Schedule II controlled substances and can constrain legal manufacturing capacity) production quota constraints, or demand increases — are not uniformly disclosed in the currently available excerpts. Clinicians seeking manufacturer-attributed reasons for a specific product should query the FDA Drug Shortages database directly, where such disclosures appear when filed.

For patients currently maintained on brand or generic amphetamine mixed salts who cannot access their usual formulation, the clinical considerations are non-trivial. Switching between immediate-release and extended-release formulations requires dosing-interval adjustments and patient counseling on onset and duration of effect. Switching to methylphenidate-class agents involves a different mechanism of action — methylphenidate primarily inhibits the reuptake of dopamine and norepinephrine, whereas amphetamines additionally promote active release — and responses do not reliably generalize across classes. Patients with a documented preferential response to amphetamine-based agents may experience suboptimal symptom control on methylphenidate, and this should be monitored explicitly rather than assumed equivalent. Non-stimulant options, including atomoxetine and extended-release viloxazine, have longer titration timelines before therapeutic effect is assessable, which has practical implications for patients mid-treatment who are switching under supply pressure rather than for clinical indication.

No definitive resolution timeline for the amphetamine mixed salts shortage appears in the available 2026 sources. Clinicians managing affected patients should document dispensing difficulties in the medical record, consider contacting their institutional pharmacy to confirm lot-level availability, and consult the FDA Drug Shortages database and the ASHP shortage list for current manufacturer-level updates. For a deeper look at the diagnosis and pharmacological management of adult ADHD, see the PsychoPharmRef post linked below.

Side-effect / Adverse-event Deep-dive

Valproate-induced hyperammonemia — recognizing encephalopathy without liver failure

psychopharmref.com · 2026-07-02

Valproate-associated hyperammonemia is a clinically underrecognized adverse effect that can occur independent of hepatotoxicity and at therapeutic serum valproate concentrations. The mechanism involves valproate's interference with the urea cycle: the drug inhibits carbamoyl phosphate synthetase I, the rate-limiting mitochondrial enzyme that initiates ammonia disposal, while simultaneously increasing renal ammonia production through its effect on glutamine metabolism. The result is ammonia accumulation that can produce encephalopathy in the absence of elevated transaminases or a toxic drug level — a pattern that routinely misleads clinicians who anchor on liver function tests as the primary safety signal.

Valproate sodium, valproic acid, and divalproex sodium all carry this risk. Topiramate (a sulfonamide-derived anticonvulsant used as adjunctive therapy in mood and seizure disorders) substantially amplifies the effect of valproate when co-prescribed; the combination has been associated with symptomatic hyperammonemic encephalopathy in case series at topiramate doses that would not be expected to cause the effect independently. Phenobarbital co-administration and high-protein nutritional states can also modify ammonia load. Patients with urea cycle enzyme deficiencies — often undiagnosed before the drug is started — face disproportionate risk, and pediatric cases have been reported at standard therapeutic doses (FDA prescribing information, date unavailable).

The estimated incidence of asymptomatic hyperammonemia during valproate therapy has been reported to be approximately 50 percent in some series, while symptomatic encephalopathy is far less frequent; however, precise population-level figures depend heavily on how aggressively ammonia is measured (Medical News Today, 2025). The clinical picture is deceptive: early findings include mild cognitive slowing, irritability, and subtle behavioral change that can be attributed to the underlying psychiatric condition or dose adjustment. More advanced presentations include asterixis, confusion, ataxia, and in severe cases, decreased level of consciousness. The AIMS (the Abnormal Involuntary Movement Scale, a structured clinician-rated instrument commonly used to screen for tardive dyskinesia) is not useful here; the relevant exam finding is asterixis, elicited by asking the patient to extend their wrists with arms outstretched and observing for the irregular flapping tremor of hepatic or metabolic encephalopathy. Serum ammonia is the critical laboratory test and should be ordered when unexplained cognitive or behavioral changes occur in any patient on valproate, regardless of liver enzyme values.

Routine ammonia monitoring is not universally mandated by current major guidelines for asymptomatic patients, but the 2026 Indian Psychiatric Society bipolar disorder guidelines reinforce the principle of systematic monitoring for metabolic complications of mood stabilizers and recommend a low threshold for investigation when clinical change is unexplained (Indian Psychiatric Society, 2026). Baseline ammonia measurement before initiating valproate in patients with a family history of unexplained encephalopathy or developmental regression is reasonable practice given the risk of undiagnosed urea cycle disorders.

When symptomatic hyperammonemia is identified, management begins with dose reduction or discontinuation of valproate, with the decision weighted by clinical severity and the indication being treated. Topiramate should be discontinued or withheld if co-prescribed. L-carnitine supplementation has been used as an adjunctive treatment given valproate's depletion of free carnitine, which participates in mitochondrial fatty acid oxidation and ammonia metabolism, though randomized evidence remains limited. Lactulose may reduce ammonia absorption in severe cases. When encephalopathy is pronounced or progressing, inpatient management with neurology input and consideration of hemodialysis is appropriate. The 2026 American Society of Clinical Psychopharmacology (ASCP) consensus statement on deprescribing psychotropics identifies adverse metabolic effects as a primary driver of regimen review and underscores the importance of structured safety monitoring when discontinuation is considered (ASCP, 2026).

For a deeper look at medication-associated blood count and metabolic abnormalities in psychiatric practice, see the PsychoPharmRef post linked below.

History of a Diagnosis

From ancient melancholia to modern nosology — how psychiatric diagnosis took shape

psychopharmref.com · 2026-07-02

Long before any formal diagnostic manual existed, clinicians and philosophers struggled to classify mental suffering. Hippocrates, writing in the fifth century BCE, described melancholia and mania as disorders rooted in bodily humors — a framework that persisted, in various guises, for nearly two millennia. The modern arc of psychiatric classification, however, begins in earnest with Emil Kraepelin (a German psychiatrist whose late-nineteenth-century textbooks introduced the foundational distinction between dementia praecox and manic-depressive insanity, a split that still echoes in today's schizophrenia–bipolar divide). Kraepelin's careful longitudinal observation of patients in Munich and Heidelberg established phenomenological description and disease course as the cornerstones of diagnosis, displacing the vague "unitary psychosis" models that had preceded him (Kendler et al., 2015).

In the United States, the first systematic attempt at counting mental illness was a single census category — "idiocy/insanity" — used in 1840. By 1917, the American Medico-Psychological Association (forerunner to the APA) had developed a more granular classification for state hospital statistics, but clinical utility remained limited (Surís et al., 2016). The aftermath of World War II, which produced vast numbers of veterans with conditions poorly captured by existing hospital-based categories, catalyzed a new effort. Drawing on the Armed Forces' wartime nomenclature and on the World Health Organization's ICD-6 (1948) — the first international classification to include a section on mental disorders — the American Psychiatric Association published DSM-I in 1952 (APA, date unavailable). Its 106 diagnostic categories leaned heavily on psychoanalytic theory, casting most conditions as "reactions" of the personality to biological, psychological, or social stressors.

DSM-II (1968) softened the "reaction" language but remained psychodynamically inflected. The pivotal break came with DSM-III in 1980, spearheaded by Robert Spitzer at Columbia, which introduced atheoretical, criterion-based definitions, multiaxial assessment, and field trials designed to improve inter-rater reliability. Spitzer's neo-Kraepelinian approach was not without controversy: critics argued that its checklist structure reified categories whose biological validity remained unproven, and that political pressures — most visibly surrounding the removal of homosexuality from DSM-II in 1973 — had exposed the fragility of diagnostic boundaries (Kendler et al., 2015; Surís et al., 2016). DSM-III-R (1987) and DSM-IV (1994) refined criteria sets and expanded the manual to nearly 300 diagnoses. DSM-5 (2013) and its text revision (DSM-5-TR, 2022) collapsed the multiaxial system, introduced dimensional measures for cross-cutting symptoms, and reorganized chapters along developmental and neurobiological lines, while the ICD-11, released by the WHO in 2019, pursued a parallel but not identical reorganization emphasizing clinical utility across global settings (Jablensky, 2009).

Active debates persist. The NIMH's Research Domain Criteria initiative challenges categorical diagnosis altogether, proposing that transdiagnostic neurocircuit dimensions may better capture psychopathology. Meanwhile, global prevalence estimates remain tethered to the very categories under dispute: the WHO estimates that roughly one in eight people worldwide lives with a mental disorder, though figures shift substantially depending on which classification system and which threshold are applied. Psychiatric nosology, as these sources remind us, is a living negotiation between clinical observation, scientific evidence, cultural context, and institutional politics — never a finished product.

For a deeper look at related diagnostic frameworks and clinical documentation, see the PsychoPharmRef posts linked below.

Found this useful? Reply to this email with your thoughts, or become a paid supporter to keep this going.

Keep reading