Sunday, November 18, 2007

Decision Making in Pediatric Psychopharmacology

Philosophy

A tool

Deficiency state, like diabetes

Return on investment unmatched

An essential utility service, with costs determined by need, plus 2%. This idea has yet to be sold.

Scandal each decade, inescapable. 1950’s: psychoanalysis; 1960’s: 90% schizophrenia diagnosis; 1970’s: hospitalizations for two years; 1980’s: recovered memory of abuse; 1990’s: “My husband’s got the bipolar;” 2000’s: left wing, PC elements; 2010’s: DSM V, a bookselling scam, and a disgrace.

Stakes

Roughly 10% of kids have a serious condition, another 10% miserable but functioning.

Conduct disorder leads to antisocial personality disorder (most crime), marked increase in all social pathologies. Each criminal commits 100’s crimes a year. ADHD plus any bad habit, catastrophic, e.g. pyromania.

Out of home placement costs $300,000 a year

Assets

Medicalization, acceptance of scientific method (verification). Psychodynamic psychiatry gone.

New medications. Direct to consumer ads.

Factors increasing treatment: zero tolerance, and PC, lawyer attacks on authority outside of central government, regulation and litigation ending punishment

Assessment

Structured interviews checklists of diagnostic criteria. Rating scales.

Target symptoms respond across diagnoses. Likely anatomically based. Anatomy = brain circuits.

Instrumental vs. affective aggression

Across settings. Difference between settings the most reliable indicator of (100%) failure to respond to meds.

Time of day variability

“Syndrome” from Sydenham (1624-1689)

Symptoms cluster. Symptoms reported by patient. Signs seen by doc.

Major: must be present

Minor: optional

Implication: patho-physiology unknown

Risk factors: predate the syndrome, e.g. familial aggregation

Associated symptoms follow syndrome onset

Course of illness. Greatest value of syndrome is reliable prediction of future course. Otherwise, a way to communicate.

Problems with Current Syndromal Diagnosis in Psychiatry

19th Century superstitions (OK for that era). Kraepelin prevails.

Final common pathways

No replication of gene mapping for most familial condition, bipolar disorder

Political and fiscal considerations, subject to corruption.

FDA requires diagnosis, no target symptom permitted.

Sequence of Common Target Symptoms in Kids

No substance abuse, including caffeine. Puberty required to get high.

ADHD – stimulants, bupropion, alpha-adrenergic agonists, atomoxetine

Anger for no reason = irritability. Aripiprazole, other atypical neuroleptics, mood stabilizers

Dangerous acts, inability to function. Same sequence as anger.

Less Common Target Symptoms in Kids

Tics = twitches. Atypical neuroleptics.

Depression = sadness for no reason, and self Injury. Anti-depressants.

Anxiety = fear for no reason; compulsions = grooming for no reason. Antidepressants in regular and in high doses.

Psychosis = auditory hallucinations, delusions. Atypical neuroleptics

Outpatient Approach

Start low, go slow, unless lonely.

Do not count on weight, because metabolism unknown

Pharmacokinetics – what the body does to the drug. Get logical blood level.

Pharmacodynamics – what the drug does to the body. Severe condition requires higher doses. Doses decrease with normalization.

Familial responses.

Always grant patient request within a class. Nothing else has a chance of working.

Prior experiences likely repeated

Niches and Verispan

Substance abuse – quetiapine, ADHD meds

ADHD – no niches; sleep in ADHD – mirtazapine, clonidine

Anger – aripiprazole if under age 30.

Psychosis – risperidone, olanzapine, ziprazidone

Depression – no niche, add mood stabilizer, including anti-psychotics. Ziprazidone has anti-depressant properties, no hunger side effect. Should have a niche in depression.

Self-injury – anti-psychotics

Anxiety – no niche

The Non-Responder – Host Factors

Adherence

Interference by other medications

Rapid metabolism (blood level) – pharmacokinetics.

Problems worsen same time of day, daily, meds running out

Problems worsen in same place but not others

Pharmacodynamics – unknowable today. Severity of illness.

Non-Response and Environment

Parental mental illness, criminality, substance abuse (no chance of improvement).

Single motherhood (diminished chance of improvement).

Non-adherence by parents

Substance abuse in adolescents (no chance of improvement).

Response to the Non-Responder

Second and third member of the class, especially if works by differing mechanism. If first helps 70%, the second will help half of the 30%, the third, half of the 15% remaining.

Add the member of another class, especially if target symptom atypical, or comorbidity accompanies

Revert to meds of the 1960’s

Clozapine for psychosis and mood disorder

Easy Psychopharmacology Mining

Off label large group tests, and in individual, on-off experiments.

Over the counter availability of low doses for mild conditions, and lower prices

End racial disparity of access

Patent term to 5 years, not 20.

Self-help to end plunder of clinical care, by land pirates. Deregulate research.

Deter single motherhood, and left wings attacks on family, school discipline, corporal punishment, other sources of authority, and structure

Demming approach to continual improvement, error reduction

Needs and Future Tools

Cool names for a) anger for no reason (not irritability); b) fearlessness disorder (opposite of anxiety)

Labs or PET scans of pharmacodynamics

First reliable gene locus

Identification of deficient neurotransmitter, location, by target symptoms.

Gene therapy

Unsure if computers beneficial. Would permit data mining by clinician.

Obstructions

Messiness

Empiricism, contradictions to evidenced based medicine

Lawyer siege from all sides: self-dealing immunity to criminals, to insurance company, to regulators, to plaintiff lawyers, to violent patients (can’t be sued, charged, nor restrained). Ban FDA.

Difficult parent, battling parents with joint custody

Guideline Problems

Parametric statistics valid to predict larger population rates

Clinical care a series of on-off (works-does not work), single case design experiments, subject to binomial distribution

Garbage science per se, will change every two years

Written by academics, organization people, devoid of credibility. Often behind by years.

Rent seeking by docs, or crushing access by payer

Both oppressive to patient and doc

Alternatives

Verispan defines the national standard of care for $5000, local std for $9000. Use reflects insider knowledge of real time patient experience.

Number Needed to Treat (NNT) or to Harm (NNH). More clinical. Allows balance of risk of illness. If outcome of illness extreme, willing to accept higher NNT, lower NNH

MD chat rooms.

Number Needed to Treat, Needed to Harm (Side Effects)

More clinically meaningful. Counts events like a clinician. Implies real cost

Corrects for rarity (like a clinician)

Usually has two comparison conditions, drug vs placebo.

High NNT, low NNH OK if untreated outcome very bad.

Research Sequence

Basic science discovery. Pharm exploitation

Clinical applications by desperate doc facing desperate patient. Most Rx off label

Clinical applications become ubiquitous

Academic submits grant proposal on ubiquitously successful application

Gets 1 in 9 approved.

Takes 4 years to complete.

Studies gathered in textbook, and by guideline makers, published two years later.

Research Applicability Problems

IRB from Fed regs, and an unapproved, uninformed, human experiment itself. Lawsuits. Make all legal research reviewable.

All non-clinical exclusion criteria violate central assumption of random selection.

Outcome measures continuous and irrelevant (scale scores)

No blood level to show any substance reached child’s brain

Inapplicability of parametric stats (continual) to on-off single case experiments described by binomial stats (like coin toss)

Doses too low due to FDA and IRB’s. Short term results.

IRB Legal Problems

Judged by competitors, with conflict of interest. (Intentional tort of interference with contract, subject to exemplary damages. To deter.)

Interference with Interstate Commerce.

Violates age discrimination, ADA.

Violation of Freedom of Association inherent in Free Speech Clause

Violates Thirteenth Amendment by interfering with adequate payment for subject participation.

Policy impact: drive out intelligent young people, decrease research, trivialize research, preclude useful negative outcomes. Telephone book thick consents useless. These are harmful effects.

There is no informed consent, not even by researcher.

IRB never considers central question of Nuremberg. Would the researcher be a subject? Exempt from review, if done.

Difficult Parents and Regulators

Others take consequences of treatment refusal.

Crazies, bullies, missing. (Total = ~5%)

Restraints prohibited.

Corporal punishment banned. Less restrictive than changing brain chemistry.

Court and lawyer intervention required, reporting for medical neglect if physical danger results.

No comments: