How To Read PI Sheets

You should take a moment to learn how to read PI sheets, as it is so not obvious as to how and when you’re supposed to read them.
Full Patient/Prescribing/Physician/Product Information Sheets OK, now your education really begins. You may or may not want or need to read the PI sheet before, during or after taking a medication. It’s up to you. These things are like the liability disclaimers in parking garages, they’re almost like contracts between the drug companies and you. The drug companies make vague promises about what the meds may or may not do, and if anything bad happens, well, they told you so. You had all of that information available before you took a single pill. And if you weren’t taking the drug right, it’s all your fault in the first place, so don’t go running to a lawyer because you’ll be wasting your time.Картинки по запросу PI Sheets of medication The paradox is that the doctors don’t want you to read these things before you take the meds, because you’ll completely freak out about the side effects or how the meds work and you’ll not want to take that drug or anything like them. And the doctors know better about these things, so don’t you dare go second-guessing them. So the beauty of the “patient information” sheets is that they are written in a language that is impossible for most patients to understand. There, you’re given all the information you need, the drug companies have a clear conscience, and the doctors don’t have to worry because you can’t understand a word of it.

Unfortunately for them, a couple people like me and Mouse became patients.

Now I’m going to help you understand these damn things, just like Mouse and I did.   We read a bunch of books, we studied web sites, we asked our doctors a bunch of annoying questions and we pestered Brian, our pharmacological consultant, with more annoying questions so we could translate these damn things into English.

Part of the problem is, there is no standard to them. There isn’t even a standard name. Patient Information? Prescribing Information? Physician Information? Pharmacy Information?   Package Insert?  Just what the fuck does PI stand for in the first place? Each drug company uses a different term. They use different formats, even within the same company the format varies from drug to drug, year to year. That’s the beauty of standards, there are so many from which to choose. I’ve experienced this sort of thing before when the bureaucracy of a large corporation collides with the bureaucracy of the Federal Government. If I ever get the money again I want to get the PDR on CD-ROM and subscribe to the official online version to see just how the hell they managed to put all that information into a database.

However, there is lots of important information contained in the PI sheets, stuff that you should know about. You should get a real PI sheet for each drug you take and you should never throw it away, even if you’ve stopped taking that drug. Why? Because I’m a geek who keeps every supporting document for his tax returns for-freaking-ever. Actually the real reason is because if you change doctors and the new doctor wants to try you on a med you’ve already tried, it’s easier to bring in the PI sheet as proof of purchase than a bottle. Plus they fit in a binder very nicely, a binder of the meds you’ve been on. A log of your side effects and how the meds worked for you. It’s all part of the strategy to take control of the healing process. By real PI sheet, I mean the ones I’ll be describing here, the ones I’ll be providing copies of, if possible. If you didn’t get one and I have a copy, download it and print it out. You should still demand a real one, but mine is better than nothing. A lot of HMOs have taken to the truly cheap-ass practice of printing out their own cheap PI sheets. THOSE DON’T COUNT. Again, they are better than nothing, and they are one reason why your co-pay is so low. If you’re stuck with one of those, ask if you can get a real one, or print out the real thing from this site. As far as proof of purchase is concerned, what you download from here doesn’t count, but the HMO sheet will. If you can’t get the real thing from your HMO, keep the one they gave you and download one from here.

How do you know if you’ve got a real PI sheet? Right after the name of the drug you’ll see the arcane glyph that represents the molecular formula of the drug. That’s the mark of a true PI sheet. Unfortunately some of those got lost in the translation to a PDF file, so even I can’t guarantee the mark of a true PI sheet for all of the ones I have.

Most of the PI sheets for the newer meds released in the US do follow a common format, so let’s go over that.

Drug Name. Brand and generic names. Sometimes multiple forms of the drug are on one PI sheet, sometimes different forms get different sheets. Don’t ask me what the requirements are that warrant different sheets, I’ve yet to figure that one out.

Description. This gets into some really arcane and obtuse stuff. We’re talking extremely precise descriptions of what this drug is, right down to its chemical composition, molecular weight and the filler ingredients.

Clinical Pharmacology. This is where you can find some of the more important information. It’s also where some of the more difficult to read information is. You’ll need a good medical dictionary to get through it. It’s broken down into different subsections.

First there’s the pharmacodynamics, which is a fancy term for how the drug works. This is sometimes called “Mechanism of Action.” You’ll often see the phrases “is presumed to” and “the precise method of action is not known.” In other words, the good people at the drug companies don’t fully understand how these meds work. Doesn’t that make you feel all warm and fuzzy. Actually they usually have a pretty good idea as to how the meds works, its just that they can’t publish that they know for sure because to really know for sure would require dissecting live humans, and I think there are some stupid laws prohibiting that sort of thing. Dumb-ass legislators are always getting in the way of science. Anyway, here you’ll learn just which neurotransmitters and receptors get messed with.  In the future I hope to render this part into somewhat understandable English.

Next up are the pharmacokinetics, which is how the drug gets from your tummy to your brain, how long it takes to do so, if it plays well with other drugs, and how long it hangs around in your system before your liver and/or kidneys get fed up with it. Fascinating stuff to be sure, but there really is stuff in here you may want to know about, most of which I’ve rendered into English for you on my drug information pages. Believe it or not, the dosages for psychiatric medications are standardized to non-smoking Caucasian males of average weight aged between 20 and 65 with no liver or kidney problems. Geez, I don’t know why that group of people is singled out for special treatment, it’s not like they run the country or anything. So if you’re not standard, i.e. you fall outside a single one of the aforementioned criterion, you may need to have your dosage adjusted. Say you’re female. Or you’re not white. Or you smoke. Any single one, you need to read this section to see if the drug companies found out about dosage adjustments. Again we run into a quirk in the reporting requirements. The FDA requires the drug companies to report any findings they have about such groups, the FDA just doesn’t require the drug companies to make extensive tests to determine if there really are any differences or not. So there may be differences, but they may not be known until a few years after the drug has been on the market. As always, we’re all guinea pigs together. Pick which sucks less, knowing more of this information at the onset and having the medications cost a lot more and take longer to be released, or having a bare minimum amount of this information be collected at the beginning and having the meds just be as expensive as they are now.

If you’re taking any medication that is difficult to discontinue, such as anything that messes with serotonin, you need to be aware of the range of the drug’s half-life. I’ve given you the mean average half-life, which is sometimes just a number pulled out of some researcher’s ass. If you’re doing a managed discontinuation and still feeling nasty effects, you may have to recalculate the withdrawal period based on a different half-life time. Work it all out with your doctor, but here is where you’re going to find the numbers with which to work.

Then there’s the clinical efficacy trials. These are the odds that the drug will work. Hey, guess what? There’s no guarantee that the med is going to work for you. And if the med is being prescribed off-label, you don’t even get odds on that. These are the odds for the intended application. Sometimes they give actual odds, in the form of a percentage of patients who responded positively, sometimes it’s just couched in vague terms that the people in the treatment group showed “significantly greater mean improvement” using one standardized test or another. What’s statistically significant you might ask? Somewhere around twenty percent. That’s right, the drug can fail for 8 people out of 10 and it can be considered worth putting on the market. So you’re not anymore crazy than your initial diagnosis if something is plain not working for you, because often the drugs just don’t work. It’s probably lawyers and marketing sleaze that prevent the drug companies from publishing the exact numbers, but the exact numbers would be tremendously helpful to doctors and patients. Then again, so would larger clinical trials, but that would just make everything slower and more expensive.

Suppose you really want to know what the odds are, and not just get a vague promise that it’s going to work in your contract after being sold the world by all those spiffy ads. Well then, what you need is the New Drug Application (NDA) data sheet. Good luck getting your hands on those. The Freedom From Information Act doesn’t make it easy to obtain them. I’ll be doing my damnedest to get my hands on every NDA I can find for a medication. When and if I can get them, they’ll be going up on this site as well.


In defense of the pharmaceutical companies, clinical trials are not the real world. Your typical clinical trial in the Los Angeles area (where many take place) pays upwards to $2,000 to people with psychiatric disorders who are not currently taking any medications, or are willing to stop taking their meds. Just whom do you think that would attract? That’s right, a high number of people who are either faking it or are truly messed up and desperate. Or both, as they may have a disorder completely inappropriate for the medication being tested. But two grand plus free check-ups for two or three months of taking some new drug? You can’t beat that with a stick. So while the double-blind methodology helps smooth out statistical aberrations caused by the fakers and the people who are taking totally wrong meds, the data are still skewed. In the trials you take one and only one med, in studies you take multiple meds, but not in the trials. In the real world you might be taking a cocktail of meds. In the trials people may not be doing much of anything else to deal with their illnesses. You, though, should be doing a lot more. Therapy, exercise, a better diet and so forth. So the low success rate in the clinical trials is understandable. Again we are faced with the dilemma, do we keep the current ‘system’ where we’re all guinea pigs, or do we have larger, longer and more extensive clinical trials that will delay the release of new meds and make them more expensive than they are now? Decisions, decisions.


Indications and Usage. This is the FDA approved usages for this medication. For those of us in Bipolarland that means you won’t see any mention of mania for a lot of the medications we use and rarely will you even see the words “bipolar” or “manic-depression.”   And the majority of those that are approved for bipolar are approved only to treat bipolar mania.  There is one, count it one drug approved to treat bipolar depression, Symbyax, and that’s nothing more than a cocktail of two previously existing drugs.  Only two medications are true “mood stabilizers” in that they deal with the mania and depression – lithium and Lamictal (lamotrigine).  Frankly Lamictal (lamotrigine) is best suited for people who are more in the Bipolar 2 part of the spectrum, while lithium works best for the classic Bipolar 1 type.  So if you don’t fit into a neat category, you’re going to need a cocktail of meds.   Many medications are prescribed off-label, that is they are used for entirely different purposes than those for which they are officially approved. They are, for the most part, perfectly safe in these applications. Sometimes in this section you’ll see an explanation of what a disorder means, sometimes not. As far as I can tell it’s a function of company policy and how generous the writer of the PI sheet felt as to how much information you’re going to get here.

Contraindications. Here you will find one of the greatest examples of legal weasel wordings in American jurisprudence. “Ubik is contraindicated in patients who have demonstrated hypersensitivity to generic ubik or any of its inactive ingredients.” Well duh. But there it is, they told you so. If you’re hypersensitive to this medication, don’t take it. It’s just like the contract on the ticket you get at a parking garage, they’re not responsible for anyone breaking into your car if your car is an obvious target for being broken into because it’s, like, a car. Any other known contraindications, usually taking MAOIs, are listed here.

Warnings. Known warnings are here. The notorious “black box” warnings are highlighted in bold print, once upon a time in a black box. Older PI sheets (not from old boxes of meds, but from old formats, as these things sometimes never get updated) still have them in a black box. These are the adverse effects that can really mess you up. Most of the time they are something that’s rather rare, or the result of something you did wrong, or from a predictable allergic reaction (as opposed to those unpredictable ones that go in the contraindications section). This is where a drug can kill you. Oh, I’m sorry, I’ve put you off of psychiatric medications for life now, haven’t I? Believe me, the death rate is really, really low. You stand a better chance of getting struck by lightening. But it’s greater than zero, so you get a warning about it, just like you do when a severe thunderstorm comes through town. Anything that is really nasty goes here. This is where far too many websites fail people and turn everyone into a cyberchondriac, they’ll tell you that a med will potentially kill you, but they won’t give you the odds. Here you get the odds. Here you can make a legitimate decision of cost/benefit and risk analysis, what are the odds of something really icky happening vs. my chances of dealing with an incredibly sucky mental illness? Sometimes with something particularly nasty this section will be right at the top of the PI sheet, after the drug name. In the vast majority of cases, even dealing with stuff that can be fatal, if you stop taking the medication you can avoid whatever it is the warning is about! Almost all of this stuff is the result of cumulative exposure to metabolites that your body couldn’t process quite right. Sudden death is extremely rare in the world of psychiatric meds, and is a concern only if you’re epileptic who experiences particularly violent tonic-clonic seizures while sleeping (you know, like I do) or with the MAOIs and a couple of the TCAs (often considered the antidepressants of last resort – but it all depends). Everything else gives you warning signs that you should stop taking a particular med before things get worse. That’s why they’re called “warnings.”

Precautions. Whenever you take a medication you have to take certain precautions. Here they are. What lab tests you’ll need, if any, and how often. How your liver and kidneys are going to hold up. Their best guess on cancer and fertility impairment. The whole pregnancy and breast feeding thing. How it works with kids and older folks. If there are any “special populations” who require dosage adjustments. As mentioned above, a special population is anyone who is not a non-smoking White male between the ages of 20 & 65. The drug-drug interactions, especially dealing with how well it gets along with other drugs in your liver. With anticonvulsants the drug-drug interactions list will run for many paragraphs as anticonvulsants wind up being extensively tested with other meds during the clinical trials. Why? Because a bunch of people are bound to have seizures. Then they’ll be given injections of whichever benzodiazepine is handy (drug-drug interaction #1) to stop it right away. Afterwards they’ll be put back on their regular meds to continue on with the trial while taking their regular meds and the new drug being evaluated. Anticonvulsants are weirder meds in a world of weird meds, so they get treated differently. The anticonvulsants also have listings for status epilepticus, which is the aforementioned spazzing out so badly that intensive and sometimes heroic intervention is required because the seizure could lead to permanent damage or death.   These super-seizures could occur before and after the trials, so there are lots of data on drug-drug interactions with anticonvulsants because everything gets thrown at status epilepticus.  And then there’s SUDEP, Sudden Unexplained Death in EPilepsy, which is the sudden death we epileptics get to worry about that is sometimes also mentioned in the Warnings section.  See, status epilepticus is the one of the expected forms of early death we live with.  There are yet unexplained deaths that may or may not be related to the anticonvulsants.  Those of you who are “only” bipolar and taking an anticonvulsant need not fret about an anticonvulsant killing you in this manner.  It’s the epilepsy that kills.

Adverse Reactions. This is it, this is the section everyone hones in on. These are the side effects. Finally we get to what everyone wants to know!

Too bad it’s not written in English.

They break up the side effects in such a way that it’s like reading the ingredients on breakfast cereals. You know how the food companies are supposed to list ingredients from the ingredient they use the most of in a product to what they use the least of. But if they use previously manufactured ingredients those items get broken down. So they’ll use a flour blend that will make it look like sugar is the fifth-most used ingredient when it is actually the second-most used. The drug companies will do a similar thing by splitting so many hairs with the side effects that it looks as if the odds are low that something in particular will happen, when actually the odds are a lot higher, it’s just that you could get various flavors of the adverse effect. Let’s take Zonegran (zonisamide) for example, as I’ve seen this particular adverse effect in action. If you look at Zonegran’s PI sheet you’ll find that you have a 6% chance each of confusion or difficulty concentrating or memory problems, a 5% chance of speech abnormalities, a 4% chance of mental slowing, a 2% chance of “Difficulties in Verbal Expression,” or a 1% each chance of incoordination or accidental injury. When in fact you stand about a 20% chance of “Me feel stooopid,” and everything else is just variations on a theme.

Just imagine if your cereal box ingredients were written in Latin. OK, it looks like the vitamins they add back into that ‘enriched’ flour are written in Latin, but that’s another issue. You look at this and it’s what the hell are they talking about? What does it all mean?

Damned if I know. I’m constantly looking up stuff. You’ve got a computer. If you don’t know what a word or phrase means, enter it into Google. You’ll learn more about it than you ever wanted to know.

And just to make things more confusing, the side effects are grouped into multiple sets. Some are in tables, some are just listed. The ones in the tables are the ones where you can see real odds and can compare the side effects of people taking the med to people getting the placebo. But those are from the clinical trials, which last 4-12 weeks. Which means a lot of side effects, like weight gain, don’t really show up, while others, like the anticholinergic effects (dry mouth, mild blurred vision, constipation and/or diarrhea, memory impairment, mild nausea) so popular with many psychiatric medications would go away after the trials end if the people kept taking the meds. Once again, can we afford longer trials? To continue, if people in the trials couldn’t put up with certain side effects you might get a table of what caused them to quit. If the drug company wasn’t sleazy about splitting up side effects as Elan did with Zonegran’s mental impairment, you’ll get a separate table breaking down all the variations, instead of just hiding the cumulative results in the lesser-read Warnings section. If a drug is approved for multiple applications, you’ll sometimes get multiple tables of adverse effects for the different clinical trials for the different clinical trials. You’re supposed to, at any rate. Whether you actually do or not is another question.

OK, that’s all the tables, that’s where you get real odds for popular side effects when you take a med for up to 12 weeks. What about taking the med for longer? And what about the unpopular side effects? This is where you see a bunch of stuff that’s just listed. Sometimes it’s just the unpopular side effects, sometimes it’s the “treatment emergent” side effects and/or “post-marketing” side effects, where it’s effects for taking the drug for longer than 8-12 weeks. Sometimes it’s an amalgamation of all of that data. What you get varies from drug to drug, PI sheet to PI sheet. There is no standard. Rather, there are plenty of standards from which to choose, so who knows to which one your drug’s PI sheet conforms. All right, these data are grouped together by types of effects, and there’s no telling how many of these you’ll see or what they’ll really be called. Here are just a sampling culled from a few different PI sheets:

Cardiovascular – Heart and vein stuff.

Hematologic/Hemic and Lymphatic – Blood stuff and your immune system. I guess they go together.

Digestive/Gastrointestinal – Tummy, guts and poop shoot

Metabolic/Nutritional/Endocrine – Weight loss, weight gain, thyroid and glandular weirdness. Sometimes Endocrine gets its own listing.

Musculoskeletal – Bones and muscles

Nervous System – Any sort of movement weirdness that doesn’t fit under muscles.   Sometimes psychiatric stuff is listed here.  Go figure.

Behavioral – You on drugs or what?

Psychiatric – Like behavioral, but where there’s a handy name for it that’s far more specific.  When not lumped in with Nervous System, of course.

Respiratory – Lungs, breathing and the like.

Reproductive – Sex problems, if not lumped together with peeing under Urogenital. Usually divided into male & female categories.

Skin and appendages – I don’t know why fingers and toes are always lumped in with your skin, but they are. This is where you look for rashes and the like.

Urogenital/Urinary – Peeing and sexual problems. Make your own jokes about it, it’s just too easy. Sometimes these will be broken down into male and female categories.

Special senses – I don’t know why they’re special, but any problems with your eyes or ears, including physical problems like glaucoma go here. Also high weirdness like losing your sense of taste, tinnitus, partial deafness and such are included here.

Body as a Whole/General/Miscellaneous – Anything that doesn’t fit in the above categories.  Sometimes there will be more than one of these on a PI sheet.  Again, your guess is as good as mine as to how someone makes a distinction.

Now you get to figure the odds by reading. Frequent side effects that happened to between 1% and 2% of patients in the trials can often become common side effects in the real world, if not more often.  It all depends on if post-marketing and treatment-emergent (i.e. anything beyond the 12-week clinical trials is included).   But infrequent effects (between 0.1% and 1%) and rare (less than 0.1%, or it happened to fewer than one person in a thousand) effects are usually going to remain infrequent and rare events. They can still happen, the odds are just against it. Unless these data include the treatment-emergent (data from studies and not just the trials) and/or post-marketing (i.e. after the drug has been released into the real world), in which case Frequent means odds better than 100-1. Most of the time these are just clinical trial data.

When people read these adverse effects, especially the fine print effects that don’t make the top-40 hit charts, they tend to totally freak out and join up with the anti-psychiatry movement. In a way I can’t say that I blame anyone who has that reaction. This is scary stuff. But let’s take an example that’s especially scary. If you look at Neurontin’s PI sheet and read all of the fine print you’ll find BREAST CANCER as an infrequent side effect under urogenital (yeah, breasts have so much to do with pissing). This is Neurontin (gabapentin) people, one of the most innocuous drugs on the planet, so the natural response is, what the fuck? Well during the clinical trials and studies, two women developed breast cancer. They had no family history, they had no exposure to other cancer-causing agents of which they were aware. Of course breast cancer happens to women with no family history and no exposure to any cancer-causing agents who don’t take any medications what so ever. But Parke-Davis is under obligation to report these events. It’s in the Warnings section as well, all 10 people who developed new tumors and the 11 who had existing ones made worse. In animal tests where rats were force-fed more Neurontin (gabapentin)  than anyone would ever take, some boy rats began to develop pancreatic tumors. So is Neurontin (gabapentin)  carcinogenic? Probably not. Neurontin (gabapentin)  had the misfortune of being overprescribed for everything. There were studies for this and studies for that. 2085 patient years. Someone was bound to get the Big C.

That’s an extreme example, but it drives home my point. A lot of the weird rare effects, including many that I’ll pull for my freaky rare side effects spotlight, will have absolutely nothing to do with the drug. Something just happened to someone in the experimental group of the clinical trial, and that person swore to God Almighty that the Zyprexa(olanzapine) gave them herpes (or maybe it made an existing outbreak worse, you just can’t tell from reading the sheet). Or made them smoke two packs a day when they told their mom they were really going to quit this year. Or made them go out drinking even if they had a 30 days sober token in their pocket. So onto the PI sheet it goes, by law.

Drug Abuse and Dependence. Most of the time there will be a sentence about how it hasn’t been evaluated in humans. Sometimes you’ll learn about how much monkeys like a particular med when compared to cocaine. Then there are meds like the benzos and the classic ADD meds, and just how you should watch yourself.

Overdosage. How much it takes and what to do about it. If known, examples will be given. Don’t try this at home.

Dosage and Administration. This is how the drug companies suggest a non-smoking White male between 20-65 (they state 18 as the low end, but it’s really 20) takes the drug for its approved application. This is frequently how you will be prescribed the med. This is not necessarily how I think you should be taking the med, though.  Sometimes it is, sometimes it’s way more aggressive a dosage schedule than I think is necessary.  In my drug guide I supply you with both my suggestions and what the drug company guidelines are.  Some of this information can get very, very complicated.

How Supplied. The size, shape, color and identifying marks of the pills and how many pills in a bottle. For liquids you get color, bottle or ampule sizes. Plus hints on storage that are more precise than “cool, dry place.”

Finally you’ll get the name and contact information of the manufacturer, copyright information of the PI sheet and often an internal document identifier.


Related Posts