Thank you for your quick reply!

Not necessarily indeed--I was merely pointing out that, by your logic, it would be equally reasonable to propose a system that allows for drugs (perhaps even only at specific doses) to exert therapeutic effects. However, given what we know about pharmacology, I see no need to introduce a generic black box into the equation.

However, it's not like these ideas aren't testable. If there is such a system, it must surely exist at the molecular level (the same level upon which drugs act), as some sort of biochemical phenomenon.

The closest biological parallel that I can think of would be VDJ recombination in immunoglobulins (which is very individual, and possess far less confounding variables than hepatic metabolism). We already know that drugs sometimes form immunogenic adducts, which can induce an inflammatory response (the hapten-carrier effect). As an academic exercise, we can hypothesize that the immune system could act as the so-called "gateway," and effectively screen for exogenous chemicals the same way that it screens for pathogens (without carriers, most drugs would simply be too small to be immunogenic). Thus, in this hypothetical, adverse events are but the byproduct of something 'getting through the filter" of the immune system.

However, it would be very difficult to make the case that a failure of the immune system is responsible for most (if not all) adverse events. After all, the hapten-carrier effect itself is a pathological one (e.g. glomerulonephritis); hardly the basis for a protective system, whose evolutionary function is to "filter out" dangerous drugs (or doses). In addition, we already know that the wide majority of drugs are cleared from the blood by the liver (enzymatic degradation and conjugation) and kidneys (filtration, secretion), rather than an immunologic mechanism (e.g. relying upon the reticuloendothelial system in either organ, which has a very limited drug metabolism capability).

The best example of a true drug "filter" phenomenon in humans would be first-pass metabolism through the liver (note that this occurs before renal filtration), which depends upon the pharmacokinetic properties of the drug in question, as well as the individual patient's characteristics (e.g. comorbidities, genetics, other drugs on-board, etc). Indeed, the hepatic portal is truly the gateway to systemic circulation for many drugs, and their activity depends upon it. However, the liver is ignorant towards what drugs work or do not work, and will just as gladly hamper a live-saving drug's bioavailability (given PO) just as it would any other exogenous chemical. Perhaps even more powerful is the recognition that a person can experience a severe, life-threatening reaction to a life-saving medication (consider penicillin). Rather than be a "generic system that vets drugs...that are too powerful," the liver takes a "shot gun" approach, and attempts to eliminate whatever it can. That's what we actually observe, rather than the selective system you're proposing.

The differential susceptibility to adverse events with certain drugs is best explained by pharmacogenomics, or by any of the mechanisms I've presented previously. I suppose my question is this: why propose a new model when the current one will suffice? Do you see an error or logical gap within our current understanding of toxicology?

To my knowledge, similarities within drug withdrawal and overdose syndromes are pharmacologically explainable [e.g. benzodiazepines do not directly activate GABA receptors (an inhibitory neurotransmitter), which prevents overdose, but allows for lethal CNS over-stimulation in withdrawal; in contrast, barbiturates can directly activate GABA receptors at high enough concentrations, which allows for an overdose, similarly to how combining alcohol (a GABA-agonist) with benzodiazepines can also allow for an overdose].

Within drug withdrawal syndromes specifically, there are characteristic differences. Consider how the withdrawal from an opiate is unpleasant, while the withdrawal from long-term alprazolam can kill. Consider the very different unpleasantness of venlafaxine (SNRI) withdrawal ("the zappies"), or that of amphetamines (hypersomnia, psychomotor slowing, etc). These syndromes are characterized by the specific pharmacology of the drugs involved.

The same can largely be said of adverse effects attributable to drugs, involving off-target effects (e.g. opioids and respiratory depression), pharmacogenetic etiologies (e.g. variations in CYP2D6 and codeine) , body position (e.g. lying down after taking alendronate), etc. Perhaps what you may be referring to is that, because many drugs are considered "foreign substances" by the body, and are present in high concentrations, they often can cause some gastrointestinal irritation. Of course, the same can be said of any enterally administrated bolus; try consuming a sugary beverage after cutting out added-sugar from your diet for a month.  This, of course, differs from pharmacodynamically unique GI adverse events; recall that commonality between antineoplastics, or the risk of NSAID-related ulceration of the stomach.

One could also note some similarities on the metabolic side of things (i.e. many drugs are cleared primarily by hepatic metabolism), though there can exist a great heterogeneity in response to drugs (different drugs are metabolized through different pathways, involving different enzymes) even within a single organ. Thus, we don't even see any unified syndrome that indicates that a drug is metabolized by the liver, though we do know that certain drugs are more toxic to the liver than others (think of APAP).

To propose a "gateway" system responsible for adverse events would be akin to proposing an overarching system that is responsible for making drugs "work." However, even within the realm of pharmacognosy (where, sometimes, the "drug" is really a synergistic mixture of many chemicals, rather than a single chemical entity), one would be remiss to suggest that a drug's activity (therapeutic or not) is somehow separate from its pharmacology. Not only does the dose a poison make, but the specific pharmacology of a poison defines the poisoning.

"To propose a "gateway" system responsible for adverse events would be akin to proposing an overarching system that is responsible for making drugs "work.""

Not necessarily. It could be a system for filtering out drugs that don't work, or work too well, or are dangerous.   A generic system that vets drugs, toxicants, poisons, chemicals, etc, and blocks those that are too powerful.

Thank you for your quick reply!

Not necessarily indeed--I was merely pointing out that, by your logic, it would be equally reasonable to propose a system that allows for drugs (perhaps even only at specific doses) to exert therapeutic effects. However, given what we know about pharmacology, I see no need to introduce a generic black box into the equation.

However, it's not like these ideas aren't testable. If there is such a system, it must surely exist at the molecular level (the same level upon which drugs act), as some sort of biochemical phenomenon.

The closest biological parallel that I can think of would be VDJ recombination in immunoglobulins (which is very individual, and possess far less confounding variables than hepatic metabolism). We already know that drugs sometimes form immunogenic adducts, which can induce an inflammatory response (the hapten-carrier effect). As an academic exercise, we can hypothesize that the immune system could act as the so-called "gateway," and effectively screen for exogenous chemicals the same way that it screens for pathogens (without carriers, most drugs would simply be too small to be immunogenic). Thus, in this hypothetical, adverse events are but the byproduct of something 'getting through the filter" of the immune system.

However, it would be very difficult to make the case that a failure of the immune system is responsible for most (if not all) adverse events. After all, the hapten-carrier effect itself is a pathological one (e.g. glomerulonephritis); hardly the basis for a protective system, whose evolutionary function is to "filter out" dangerous drugs (or doses). In addition, we already know that the wide majority of drugs are cleared from the blood by the liver (enzymatic degradation and conjugation) and kidneys (filtration, secretion), rather than an immunologic mechanism (e.g. relying upon the reticuloendothelial system in either organ, which has a very limited drug metabolism capability).

The best example of a true drug "filter" phenomenon in humans would be first-pass metabolism through the liver (note that this occurs before renal filtration), which depends upon the pharmacokinetic properties of the drug in question, as well as the individual patient's characteristics (e.g. comorbidities, genetics, other drugs on-board, etc). Indeed, the hepatic portal is truly the gateway to systemic circulation for many drugs, and their activity depends upon it. However, the liver is ignorant towards what drugs work or do not work, and will just as gladly hamper a live-saving drug's bioavailability (given PO) just as it would any other exogenous chemical. Perhaps even more powerful is the recognition that a person can experience a severe, life-threatening reaction to a life-saving medication (consider penicillin). Rather than be a "generic system that vets drugs...that are too powerful," the liver takes a "shot gun" approach, and attempts to eliminate whatever it can. That's what we actually observe, rather than the selective system you're proposing.

The differential susceptibility to adverse events with certain drugs is best explained by pharmacogenomics, or by any of the mechanisms I've presented previously. I suppose my question is this: why propose a new model when the current one will suffice? Do you see an error or logical gap within our current understanding of toxicology?

"Perhaps what you may be referring to is that, because many drugs are considered "foreign substances" by the body, and are present in high concentrations, they often can cause some gastrointestinal irritation."

I do indeed wonder if there is an excess of "gastrointestinal" drug side effects.   However, I have a problem with the interpretation of this.   Could someone please give the official definition of "Gastrointestinal complaints", and say how they think the symptoms listed relate to the GI tract? 

"Within drug withdrawal syndromes specifically, there are characteristic differences. Consider how the withdrawal from an opiate is unpleasant, while the withdrawal from long-term alprazolam can kill... These syndromes are characterized by the specific pharmacology of the drugs involved.  The same can largely be said of adverse effects attributable to drugs" 

Yes, but this should not blind us to the fact that there might be a common reaction as well.   Surely the similarity in withdrawal symptoms between so many different drugs precludes any pharmacological mechanism specific to any drug or class of drugs?

"However, given what we know about pharmacology, I see no need to introduce a generic black box into the equation. However, it's not like these ideas aren't testable... why propose a new model when the current one will suffice? Do you see an error or logical gap within our current understanding of toxicology?"

Absolutely, I think toxicology does not see the wood for the trees, which is why I put up this question.   While we are thinking about life in the jungle, ancient man would not have survived very long there without an early warning Substance Sentinel System.   Ingesting some potential food or toxin and then waiting to see if it poisoned you is not a sensible evolutionary strategy.  Many modern drugs originated from the jungle.

Some chemicals irritate the skin or mucous membranes and some will irritate the gastrointestinal tract. You only need to experiment with a small amount of these to learn what may make you sick. Some chemicals interact with our nervous systemn or hormones and this can make them beneficial or toxic (depending largely on dose). And some (such as arsenic) are going to poison you if they are present in anything but trace amounts.

Your sense of taste and smell do act as "sentinels", in the sense that we find certain flavors and smells unpleasant, but there is no universal sentinel systemn.

A good example is that many alkaloids are intensely bitter, and that within the human population there is a range of sensitivity to bitter compounds. Some of us like bitter flavors more than others, and this makes good evolutionary sense; some of the population are more cautious, some more adventurous, with novel plants that may be a valuable food stuff, a medicine, or may be toxic. Trial and error is how human populations have learned to exploit new environments.

Here in Western Australia there is a cycad called a Zamia Palm. https://en.wikipedia.org/wiki/Macrozamia_riedlei

The large, attractive seeds are intensely poisonous, but Aboriginal people have learned to prepare them for eating by husking then grinding them into a flour-like powder, then bleaching, rinsing, bleaching and rinsing that powder repeatedly. It's a very involved process, but at some stage in pre-history people were hungry enough to learn how to do it. People who live in intimate association with their environment for many generations pass this sort of knowledge (that must have been acquired once-upon-a-time through trial and error) to their young folk via their culture. 

When I went through wilderness survival training we were taught to assess potential foodstuffs from unfamiliar plants. There are some indicators (colour of fruits, type of sap) that are warning signs, and then (as a last ditch alternative to starvation) a potential food plant is tested as follows; a small piece is wiped on the skin and then you wait to see if there is any inflammation or irritation. If not, you place a bit on the outside of the lip and again observe if there are any symptoms. If not, you repeat this process on the inside of the lip, then against the buccal mucosa. If this causes no discomfort you ingest a small amount and wait.

In relation to your original question, some side effects are a direct result of toxicity and as Joseph has explained these are clearly related to the actions of the neurotransmitters or hormones that the specific substance affects. For example an opioid overdose shares no common symptoms at all with an episode of amphetamine toxicity.

Withdrawal symptoms are not a "side effect" in this sense, rather they are symptoms of neuroadaption to the constant presence of the drug in question. To some extent withdrawal symptoms are "learned" (conditioned) responses. The withdrawal profile of different classes of drugs differ, but there are some withdrawal symptoms that are common to many drugs. These more common symptoms are very similar to what animal researchers call "sickness behaviours" and they are probably related to immune/inflammatory responses. https://en.wikipedia.org/wiki/Sickness_behavior

Regards,

Paul.

"Some chemicals irritate the skin or mucous membranes and some will irritate the gastrointestinal tract. You only need to experiment with a small amount of these to learn what may make you sick... When I went through wilderness survival training we were taught to assess potential foodstuffs from unfamiliar plants... a potential food plant is tested as follows; a small piece is wiped on the skin and then you wait to see if there is any inflammation or irritation..."

   Does this explain why there seems to be an excess of dermatological side effects in Meyler's SED, even when the drugs are taken internally? 

If you look back to the original "lock and key" idea of small molecules binding to proteins, then yes, molecules of similar structure would be expected to have similar side effects.  There ARE specific toxicities to many chemicals, these are liver or kidney damage.  Hopefully, but not entirely impossible, is a person that can't metabolize excrete a drug like most people and would end up in serious trouble with either of these.

Beyond that, look at the prescribing information.  Clinical studies for NDA's typically include thousands of subjects, so most, if not all of the side effects are more or less known.   Also, beware when death is listed as a possible outcome.  What likely happened is that single trial subject died for unrelated reasons, but the death has to be reported in the trial outcomes.

Drug side-effects, much like mechanism of action, is a very interesting area of pharmacology.  The different isoforms of drug targets that different people possess can change the efficacy as well as the side effect profile.

One size has never fit all.

Best,

John A. Hatten

From a quick scan of the above argument, I could say that there is a common pathway for all drugs to go through; namely, hepatic metabolism, which render any drug molecule to be an active or inactive. Therefor, we can see different effects either desirable or harmful resulting from the metabolism of certain drug groups.

These are two interesting responses worth investigating. On the whole, serious adverse events, which are quite common, depend on the drug, the disease, and the host. Risks are 1 in 5 new drugs under full review and 1 in 3 when regulators do accelerated reviews, which are increasingly common. See details and references in http://www.pharmamyths.net/institutional_corruption_and_the_myth_of_safe__effective_drugs_122645.htm.  

And http://www.pharmamyths.net/a_trio_of_short_critiques_of_the_fda_127980.htm 

Over time, the patterns on development, biased testing, production of biased knowledge, and medical practices based on it have led to drugs being the 4th leading cause of death. See http://www.pharmamyths.net/the_risk_proliferation_syndrome_behind_prescription_drugs_127198.htm

Distinguishing foods from toxins and poisons has been important in evolution. It might be thought that this depended on specific mechanisms, eg avoiding food with bitter tastes.  However, the abstract below confirms that there is a strong general factor in food preference, hence drug monitoring is likely to have similar architecture:

"Am J Clin Nutr. 2016 Apr;103(4):1099-104. doi: 10.3945/ajcn.115.122945. Epub 2016 Feb 10.

Common genetic architecture underlying young children's food fussiness and liking for vegetables and fruit.

Fildes A..

BACKGROUND:

Food fussiness (FF) is common in early childhood and is often associated with the rejection of nutrient-dense foods such as vegetables and fruit. FF and liking for vegetables and fruit are likely all heritable phenotypes; the genetic influence underlying FF may explain the observed genetic influence on liking for vegetables and fruit. Twin analyses make it possible to get a broad-based estimate of the extent of the shared genetic influence that underlies these traits.

OBJECTIVE:

We quantified the extent of the shared genetic influence that underlies FF and liking for vegetables and fruit in early childhood with the use of a twin design.

DESIGN:

Data were from the Gemini cohort, which is a population-based sample of twins born in England and Wales in 2007. Parents of 3-y-old twins (n= 1330 pairs) completed questionnaire measures of their children's food preferences (liking for vegetables and fruit) and the FF scale from the Children's Eating Behavior Questionnaire. Multivariate quantitative genetic modeling was used to estimate common genetic influences that underlie FF and liking for vegetables and fruit.

RESULTS:

Genetic correlations were significant and moderate to large in size between FF and liking for both vegetables (-0.65) and fruit (-0.43), which indicated that a substantial proportion of the genes that influence FF also influence liking. Common genes that underlie FF and liking for vegetables and fruit largely explained the observed phenotypic correlations between them (68-70%).

CONCLUSIONS:

FF and liking for fruit and vegetables in young children share a large proportion of common genetic factors. The genetic influence on FF may determine why fussy children typically reject fruit and vegetables."

not to mention the color, taset, smell definitly affect the acceptance of food or medications.

Not all drugs are eliminated through the liver; there are a variety of other avenues available for the metabolism and excretion of exogenous chemicals. Even for those drugs that are eliminated hepatically, this too is an oversimplification, for the liver is a biochemically complex organ with an array of enzymatic processes at its disposal. Further, some drugs are actually activated through hepatic metabolism (some pro-drugs). The wide spectrum of drug toxicity (and activity) cannot be explained through an understanding of the liver alone, or the oversimplification of the liver as a "gateway" for screening for those chemicals which the body  recognizes as "beneficial." The body does not know the difference between a chemical that is intended to save its life, vs. one that will kill it (ask an asthmatic); most everything is based upon a comparison of self vs. non-self (which is why cancer, a disease of the self, is so difficult to treat; cancer cells look very similar to the body's own healthy cells, which allows it to evade the immune system). 

While there may be a strong genetic factor in food preferences, and the role of a medication's presentation (pharmaceutical elegance) has been well documented, I don't see how this relates to the hypothesis that there necessarily exists a system through which to give preference to (or somehow select for, at the cellular level) beneficial drugs (or adequate doses). As you are aware, the physical properties of medications have been altered to make them appealing; I doubt that anyone would want to drink APAP if it was dissolved in water, rather than cherry syrup, even if they had a fever.

"The wide spectrum of drug toxicity (and activity) cannot be explained through an understanding of the liver alone, or the oversimplification of the liver as a "gateway" for screening for those chemicals which the body  recognizes as "beneficial."

The main reason for discounting the liver as a candidate for an early warning system is that by the time the liver gets to work the poison is in the body and has already done its dirty work.

In my opinion exists spesific side effects for each drug and at the same time general effects for the majority of drugs

"The main reason for discounting the liver as a candidate for an early warning system is that by the time the liver gets to work the poison is in the body and has already done its dirty work."

That's only for topical GI poisons, however. All enterally administrated drugs must pass through the hepatic portal to enter systemic circulation. It's just that there may not always be a significant first pass effect. The wider spectrum of adverse drug events cannot be explained through topical, local effects alone.

With yesterday's Mexican president's law of allowing marijuana's consumption, it's more legalised. 

I am amazed at the extremely great variability we see in the effects of almost all drugs. They all usually have several effects. We look at these and decide if we like one or more of them. Then the drug is marketed for that effect and the other responses are called side effects. You might look at Jelliffe R: Concerning Antiplatelet Therapy. A letter to the editor, Am. J. Cardiol., 2014;113:2086-2087). The suppression of individual data that has to be there and their use of only the average antiplatelet response versus outcome response is, in my opinion, a most unethical marketing approach, telling us the drugs do not need to be monitored, Monitoring is an opportunity to optimize therapy, and is definitely NOT a hassle, as it is advertised to be. Notice that the authors did not answer my question about who designed the study, but rather chose again only to answer wno funded it.

We could develop a target goal of antiplatelet dosage with individualized regimens and monitoring to assure achievement of that desired best balance between poor outcome versus bleeding. Why do they never seem to look to see who bleeds and who has further bad outcomes?  That is why I still prefer Coumadin. You can monitor it and dose the patient to the best effect, usually at about an INR of 2.0.

I think most drugs have multiple effects. We choose to call one a main effect and then the others are simply called side effects.Look at the example of Viagra.

Greetings colleagues! I agree with Dr.Gordon that the general effects are common to several commonly used medications (eg., cough when taking ACE inhibitors, uletserogenic effect of nonsteroidal anti-inflammatory agents). But in addition, there are idiosyncrasies reaction. They are very specific and unpredictable.

Sincerely, Zmicer

"Your sense of taste and smell do act as "sentinels", in the sense that we find certain flavors and smells unpleasant, but there is no universal sentinel system..."

   I think there is indeed a Substance Sentinel System located in the vestibular labyrinth.  Why there, one might well ask?  The inner ear detects sensory changes via hair cells, which rely on very finely adjusted metabolic and ionic concentrations in different cochlear fluids, so my guess is that it is the part of the body most sensitive to abnormal substances in the blood.  On detecting anything out of the normal, the vomiting reflex is initiated, immediately expelling harmful stomach contents.

   One can see this system working in chemotherapy with its common and unpleasant side effects.  Powerful medicines are taken to destroy the cancer cells, so the ear works overtime generating nausea and vomiting in an effort to counter these drugs.  Another symptom of an overstressed inner ear is vertigo or dizziness, which is why I think these are greatly overrepresented as drug side effects in Meyler.

See below for another example from Schiz Bull of a strong but under-researched general factor.   Many drugs have been shown to trigger psychosis, and as they have widely differing physiological modes of action, the simple explanation is that they are all setting off a common general reaction, as flagged by the general factor, as well as their specific effects.

***************************************************************************

"Testing the Psychopathology of Psychosis: Evidence for a General Psychosis Dimension

Ulrich Reininghaus, Stefan Priebe and Richard P. Bentall

Accepted November 11, 2011.

Abstract

Background: Psychiatric taxonomists have sometimes argued for a unitary psychosis syndrome and sometimes for a pentagonal model, including 5 diagnostic constructs of positive symptoms, negative symptoms, cognitive disorganization, mania, and depression. This continues to be debated in preparation for impending revisions of the Diagnostic and Statistical Manual of Mental Disorders and the International Classification of Diseases. We aimed to identify general and specific dimensions underlying psychopathological features of psychosis.

Methods: The samples comprised 309 patients admitted to psychiatric services in the acute phase of their first or second episode of psychosis and 507 patients with enduring psychosis recruited from community mental health teams. Patients’ symptoms were assessed on the Positive and Negative Symptom Scale. Analyses compared unitary, pentagonal, and bifactor models of psychosis.

Results: In both samples, a bifactor model including 1 general psychosis factor and, independently, 5 specific factors of positive symptoms, negative symptoms, disorganization, mania, and depression gave the best fit. Scores of general and specific symptom dimensions were differentially associated with phase of illness, diagnosis, social functioning, insight, and neurocognitive functioning.

Conclusions: The findings provide strong evidence for a general psychosis dimension in both early and enduring psychosis. Findings further allowed for independent formation of specific symptom dimensions. This may inform the current debate about revised classification systems of psychosis."

From Nalivaiko et al in Temperature 2014;1;164

"Our bodies possess several lines of defense against intoxication. The first level is distant – unpleasant smell or unappealing appearance of the food would prevent us from its ingestion. The second level is represented by gustatory receptors – we spit out anything with nasty taste. Level 3 comprises the protective mechanisms in the stomach which is vomiting (including 5-HT3 receptors on the afferent vagal ending that, when activated, cause nausea and vomiting). If a neurotoxin passes this line of defense, it may then activate nausea/vomiting and hypothermia by acting in certain “sensor” brain areas, e.g., area postrema – a fourth line of defense. Taking into account Reason and Brands’ sensory mismatch theory of motion sickness, Treisman made 2 suggestions: i) that another “sensor” area comprises “the systems involved in controlling movement, including eye movements, and determining the location of the body in space” that are “almost continually in action and highly susceptible to even a minor degree of disruption; they constitute an ideal warning system for detecting early central effects of neurotoxins, where these have not activated more basic levels of defense;" and ii) that stimuli that elicit motion sickness, just by accident, activate this last level of defense. In other words, vestibular and/or visual stimuli capable to provoke motion sickness do so by accidentally activating integrated response primarily designed to attenuate effects of toxins (by reducing metabolism) and to prevent their ingestion in the future (by inducing nausea that is extremely efficient in producing aversive conditioning). Speaking about nausea, Treisman concludes: “If this suggestion is correct, motion sickness is an adaptive response evoked by an inappropriate stimulus”; we believe this statement is equally applicable to MS-induced hypothermia."

****************************************************************************

Comment:    I think Level 3 is in the inner ear, not the stomach.   This is much simpler than the offered alternative scenario.  There are, I think, an excessive number of symptoms of motion sickness listed in Side Effects of Drugs, consistent with this. 

The answer is Yes for both cases. Some side effects are nonspecific and common for all drugs, as an example allergy, etc. other are specific for each group of drugs, as an example NSAID -impact on gastric mucosa, etc.

An example of how drugs preferentially go for the ear (unless anyone is willing to say Jose has a weak brain or mind?):

"For all the drama, it was clearly not a life-and-death scenario and I didn't want to get in the way. Later, I didn't need to hunt Mourinho down. He found me. He came across smiling but a little bit sheepish and embarrassed.

The gist of his explanation was this: 'What you saw was nothing serious.

'I went all dizzy and fell to the floor as I was walking to my room but everything is OK. 

'What's happened is a mix-up. I have been taking some pills for a bad back. Then on tour I've started losing my voice because I have been doing so much shouting and organising in training.

'I saw the doctor and was given some drugs to treat my throat. Unfortunately the pills for my back and the drugs for my throat have reacted to each other and made me disoriented.

'It has all been sorted out now and I'm fine.'

Read more: http://www.dailymail.co.uk/sport/football/article-3802905/Jose-Mourinho-collapsed-hotel-floor-Chelsea-s-USA-tour-Dr-Eva-Carneiro-help-him.html#ixzz4LSmIyJq2

Follow us: @MailOnline on Twitter | DailyMail on Facebook

you also may want to consider how data on "side effects" are gathered. Questionnaires may be open-ended or ask about specific symptoms. Also consider the difference between the investigational drug and the comparator.

In my experience, practically all studies report headache and gastrointestinal disturbances as adverse effects - also for a placebo arm of a study.. I see no pattern in true drug-related adverse effects, which occur more frequently than with the comparator/placebo. If you really want to get involves in thsi fascinating topic, do look at the literature on nocebo effects. 

"In my experience, practically all studies report headache and gastrointestinal disturbances as adverse effects"

This confirms my impression that these adverse effects are disproportionately reported as occurring over and above the placebo arm.   For gastrointestinal effects, read nausea, which in my opinion (and Erasmus Darwin's) never arises directly from the stomach or GI tract, it is a symptom of vestibular dysfunction, usually in the inner ear (qv sea-sickness).  A recent paper (? latest issue Otol Neurol) points out that Meniere emphasised headaches when describing his syndrome.

Anthony, allow me to clarify. What I meant is that some events, such as headache and gastrointestinal disturbances, are quite common - whether you are taking an investigational drug or not. If you are, you are likely to attribute the occurrence to the drug. There are actually studies looking into which symptoms are likely to be reported with placebos. I did not wish to imply that any given adverse effect is more common with all active substances in general.

Here is an example of a specific long-term effect of a drug (nicotine) overriding the short term unpleasant generic effect from activation of the body's toxin sentinel system in the inner ear by his first cigarette:

"I saw sparks and stars, went enjoyably giddy, horribly faint and then violently threw up. It was the stat of a tempestuous love affair that lasted, on and off, for the next 11 years."

Charlie Gilmour Evening Standard Oct 5 2016 p15.

The following old observation is consistent with the idea that there is a general non-specific action of drugs on the inner ear ("the auditory nerve apparatus").  The most obvious difference between the ear and other sense organs is the presence of exquisitely adjusted fluid compartments:

"It is also well known that the auditory nerve is more often affected than any of the other nerves of special sense by the action of drugs circulating in the blood.  One must only take into consideration the action of such medicaments as quinine, salicylic acid, morphine, chloroform, tobacco, alcohol (Alt), and the oil of chenopodium, which exercise a temporary, often, however, a permanent, influence on the function of the auditory nerve..."

A Politzer Diseases of the Ear 5th Ed 1909 

Another example supporting Politzer (see last entry):

"Though I tried -- and honestly made the effort as a teenager -- I never got the hang of smoking.  Alcohol hit the spot like magic ointment but cigarettes made me nauseous..."

Carol Midgley London Times Nov 18 2016 p28

"I think there is indeed a Substance Sentinel System located in the vestibular labyrinth."

More evidence for a general factor representing a hypersensitive inner ear:

"Reports of central nervous system (CNS) effects of ibuprofen appear to be common at therapeutic dosages"

KD Rainsford (2015) Ibuprofen p 478 

************************************************

Common complaints were headache, vertigo and tinnitus, more parsimoniously regarded as an inner ear, not CNS, syndrome.