Consumers Report More Adverse Events Than Docs
89 CommentsBy Ed Silverman // June 11th, 2008 // 8:32 am
Consumers became the major source of adverse event reports to the FDA in 2006, replacing physicians, FDA News reports. In 1998, consumers reported 23,691 adverse events, compared with 48,314 physician reports. By 2006, a tipping point occurred - there were 127,475 consumer reports versus 113,444 physician reports, according to FDA data.
And last year, the gulf grew wider - consumers reported 174,216 adverse events in 2007, which was 53,216 more than reported by docs, according to the data.
Why the shift? FDA News speculates that consumers may have greater awareness of safety issues. That seems to be a reasonable assumption, given the headlines over antidepressants and Vioxx in 2004, a trend that continued last year with Avandia and Chantix. However, we would point out that safety issues made headlines, albeit sporadically, in previous years in connection with other drugs.
One limitation in examining the data is that some adverse event submissions do not have the reporter listed on the form, or there are multiple reporters for one adverse event, according to FDA News. For example, in 2007, there were 397,902 instances of an identifiable reporter for an adverse event. However, the FDA entered a total of 364,449 direct, 15-day and periodic reports into the adverse reporting system during that same year, a discrepancy of more than 30,000, FDA News writes.
The FDA also posted adverse event report data from other healthcare professionals, which show that the role of pharmacists in reporting adverse events has diminished. While total adverse events entered into the agency’s reporting system more than tripled from 106,384 in 1998 to 376,902 in 2007, pharmacist reports only increased from 18,193 to 21,343 during that same time frame.
Tom
FDA News doesn’t fully explain how the system really works. Relatively few consumers make adverse event reports directly to the FDA. Most of the time, they contact pharma companies to ask medical questions, and in the course of these discussions, often mention a medical problem they’ve experienced while on the drug. Companies must treat these as adverse events and report them to the FDA, even if there’s no evidence of a drug link (for example, a life-long smoker who is diagnosed with advanced lung cancer after a week on a nicotine patch is reported to the FDA by as a patient with a serious adverse event). What appears to be a spike in voluntary patient “reporting” is most likely the by-product of a demand for medical information, caused by growing consumer interest in health topics, including the drugs they take, reduction in time that health professionals spend counseling patients, direct to consumer ads by drug companies, and media outlets turning any hint of a new safety issue into evidence of a vast global conspiracy. Reporter identity is indeed a data limitation with consumer adverse event reports, as mentioned in the article. The greater limitation is that it’s often impossible to tell if an individual report is related to the suspect drug, to some other drug or supplement the patient is taking, to his or her state of health, to genetics, environmental factors, etc. The value of individual adverse events is that, collectively, they can give the FDA a signal that an issue needs to be studied in more detail with epidemiology studies or some other methodology. That scientific principle is often conveniently ignored by gadflies and journalists who tout specific numbers of adverse event reports as definitive proof of a safety issue.
Jack2
I view patient’s growing demand for medical information as a positive. But well said Tom.
Ed Silverman
Hi Tom,
Thanks for the added insight. I wonder at what point that DTC contributed to this transformation, given that the advertising started 10 years ago. In any event, for the record, I wasn’t trying to make any particular point about how side effect numbers should be examined before drawing conclusions. The message here seemed to be greater consumer awareness.
Anyway, thanks for writing in,
ed
Jack2
The post doesn’t mention DTC. I speculate DTC didn’t cause this - the internet caused it. It’s now pretty easy to get a company’s phone number, and consumer’s use it to call with questions.
Bruce
Tom,
Do pharma companies in the US have to report all reported AEs or do they have a judgement call based upon your scenario? The reason I ask is that in Canada, pharma is legally responsible to report any or all AEs provided, whether they are spontaneous or requested.
Bruce
Atlex
Bruce,
Pharma companies are required to report all AEs. Moreover, in my company (and most others)any employee who receives a information about an adverse event is required to report on it. For instance, I was at a holiday party and a friend mentoned a possible side effect to one of the drugs marketed by my company. By company policy, I am required to report this. It would then go into the database that goes to the FDA.
Atlex
Nathan
I can vouch for Altex’s statement. During our employee orientation at my company EVERY EMPLOYEE (janitor to exec) is required to learn how to submit AE reports to company officials. We were told to report anything that we learned from family, friends, etc whether in or out of the workplace environment. There is no judgement call. The FDA makes the judgement calls. We simply report what we hear and see.
Jaynesday
Hi All,
I also thank Tom for his insight but there seems to be a too much speculation about AE reporting, begining with the FDA unfortunately. Someone needs to get down to the brass tacks on the issue soon. I’ve heard too many “most likely” statements on who reports, why they report, what’s reported, the conclusions of reports, the validity of the reports, on and on.
To paraphase - “It’s *most likely* we’re talking about *some number* of serious reactions to drugs and devices” - just doesn’t cut the mustard.
Per Tom - “The value of individual adverse events is that, collectively, they *can give* the FDA a signal that an issue needs to be studied in more detail with epidemiology studies or some other methodology.”
I would like to know if the “can give” part really does give. “My feeling” is that it doesn’t but who really knows?
Lauren
I, the mother, reported my son’s death from Zyprexa to Medwatch, being sure that the doctor would not, nor would the people at the intensive care unit. I also reported my daughter’s reaction of suicidality from Lamictal and was told by the FDA to tell this to Glaxo.
It has always been my opinion that this reporting, and the Medwatch System, are ill-used or seldom used, no matter how many people report adverse effects. It is amusing to me that all the Pharma people here self righteously say that they are all ordered to report an adverse effect when it is their own companies that are so often responsible for hiding the evidence of lethal effects from others in the company, federal regulators, and saddest of all, the victims.
Doug Bremner MD
I think that what we are seeing here is that individual healthcare consumers are no longer content to sit back and passively accept what comes their way in terms of healthcare. The rise of internet sites like medications.com has given them a tool to educate themselves and bypass the traditional sources of information.
Tom said “The value of individual adverse events is that, collectively, they can give the FDA a signal that an issue needs to be studied in more detail with epidemiology studies or some other methodology.”
Actually not true. If you look at textbooks of pharmacoepidemiology “epidemiology” is not listed as one of the factors used to establish causality of a side effect being related to a drug. Nor is conducting a clinical trial, which is neither ethical nor feasible. Individual case reports are, however, used, as are other factors including rechallenge with recurrence of the side effect.
Bruce
Atlex/Nathan,
That was the same way in Canada. I assume that it is due to the fact that Health Canada and FDA share their databases they would want consistent reporting methods.
B
Jack2
How would you improve the system Doug?
Doug Bremner MD
As for healthcare consumers I think medicine / the body should be taught in school so that people have the basic framework to educate themselves about prescription drugs and other relevant issues. I mean your health is just as important as the history of ancient China, right? And we don’t teach anything in school. As for unanticipated side effects and adverse effects for prescription medications that come out after approval I think there should be a government agency separate from the FDA to review data and provide real surveillance. With PDUFA drug surveillance has been pretty much scrapped. Talk to anyone at the FDA and they will tell you off the record that they have no ability to do drug surveillance. It’s pretty much a crap shoot for consumers and with preemption on the horizon the last check (i.e. the courts) will be gone too. Maybe Just in Mich can give us a little lecture about that.
Jack2
Although you will never have all of these, the criteria for establishment of causality:
1) Coherence with existing information (biological plausibility)
2) Consistency of the association
3) Time sequence
4) Specificity of the association
5) Strenght of the association
a) Quantitative strength
b) Dose-response relationship
c) Study design
For individual spontaneously reported AEs there’s no effort to ensure that criteria 1, 2, 3*, 4, 5a, 5b, or 5c are met. We just collect all AEs. When looking at a group of spontaneously reported AEs together you may be able to improve the situation, and find evidence for some of these things (5b maybe, 5a maybe) - although you’ll never get very good information as to the frequency of the AE, which is the second question everyone has after causality. This mostly comes from the textbook Pharmacoepidemiology 3E.
This book also lists study desings in terms of evidence grades, and the epidemiology evidence grade list mirrors other lists:
1. Randomized clinical trial
2. Cohort study
3. Case-control study
4. Case series
5. Case reports
AEs essentially amount to case reports, and furthermore they amount to case reports not necessarily reported by an HCP, and not necessarily reported by someone who actually thinks the drug caused by the AE.
I don’t think that collecting spontaneously reported AEs has no value. Again, going back to what Tom said, they can give a signal that an issue needs to be studied in more detail with epidemiology studies or some other methodology. I would also add that spontaneously reported AEs can never give an estimate of frequency.
*there is a time effect in that the person reporting the AE is on the drug, but it is extremely non-rigorous for AEs.
Doug Bremner
Time effect meaning that the side effect occurred after the drug was started. I would also add challenge-rechallenge (side effect goes away when off the drug, and comes back when you go back on) which was cited in the textbook pharmacoepidemiology as being sufficient in itself as support for causality, and class effect (other drugs in the same class cause similar side effect). You can’t do a randomized clinical trial to determine if Chantix causes suicidality: it is neither ethical or feasible. This is one example but I could go on and on. In the absence of this you need a method to assess causal relationships between drugs and adverse events. Healthcare consumers would not (and should not) accept the fact that in the absence of a randomized clinical trial or an “epidemiological” study that there is no evidence that a drug is associated with an adverse event. The textbooks on pharmacoepidemiology don’t use this logic, why should they.
laura
For what it’s worth, here is my family’s experience with the FDA MAUDE database…
My mother died from an adverse reaction to a medical device. When she was in the hospital, the doctors and the hospital did not seem to have consulted the database because we were told that only 3 other people had reacted to the medical device in the way that she had reacted. After she passed away, we called the FDA and were told of the MAUDE database. We found that there were many reports concerning this medical device, most of which match the reactions that my mother endured. As of today, there have been 271 reports on the MAUDE database concerning this device…271 reports concerning this device and no changes to the labeling or removal of the medical device from the market. There have been several deaths reported, although, other than from my family’s own report, my mother’s is not one of them. The medical device company has never reported her reaction as a death, it is reported as “required intervention”. I believe that if a salesperson reports that a person has suffered an adverse event, it is indicated in the MAUDE report. That, also, has not happened in my mother’s case. I have to think that my mother’s doctor would have reported this information to the sales rep, given the fact that he said he would never use the medical device again.
I guess the bottom line, as far as I’m concerned, is this. Physicians don’t use the database to access information about adverse events, the FDA does not use the database to monitor medical devices and sales reps do not use the database consistently to report adverse events.
This database was designed as a way for the FDA, medical professionals and consumers to monitor medical device adverse events. It appears, in my mother’s case, that it has failed.
Jack2
But you also can’t say Chantix causes suicidality.
1. We don’t know what % of total events are actually captured as reported AEs.
2. We don’t know the *background* rate of this behavior in this patient population.
3. Without randomization, we don’t know if the population on Chantix differs from the population not on Chantix (example - the population on Chantix may be more motivated to quit. example - they could be less likely to *cheat* while trying to quit. Would this affect other behaviors - I have no idea).
I’m not saying I have the answer. I’m not saying spontaneously reported AEs are worthless. Just that I disagree with the assertion that spontaneously reported AEs should trump data from a randomized clinical trial, where we make a strong effort to keep populations identical and we make a strong effort not to miss AEs in the placebo and active groups.
Justice in Michigan
For Atex, Nathan, Jack 2 and others directly involve in industry.
I appreciate the legal requirements to report serious AE’s that come to the attention of companies.
But what we’ve learned over the years is that it is quite possible to be in compliance in this regard, without that necessarily meaning much. For example:
1. Company reports AE’s in a way that goes out of its way _not_ to flag connections that company believes are important. There are many such examples. In fenphen, for example, the valvular issues were essentially buried in company-submitted reports. So that it was up to FDA’s Office of Drug Safety to find them. Not surprising, they didn’t until Mayo Clinic news conference and some whistle-blowing.
2. Company does its own research on AE’s submitted to Medwatch (through FOIA to FDA) which suggest serious problem with their product. The reports have been filed, but company does not submit its own summary of their implications, which do not look good. Indeed, in promotional materials, company goes out of its way to suggest precisely the opposite of what it knows AE reports suggest.
3. Company considers doing independent study based on Medwatch reports, but opts not to. In the now famous phrase from the Bayer/Baycol case, “If FDA asks for it, we will have to give.”
This is a serious question. It seems to me such examples suggest that companies can, with relative ease, be in compliance with FDA reporting requirements yet still go out of their way to suppress drug safety information.
Do you feel it is company’s responsibility to go further, or to wait for FDA to find whatever there is to find - essentially taking a cat-and-mouse approach to regulation?
Would you be personally comfortable if you knew your own company took an approach as described as one of the three examples above?
HorusCat
Lauren,
Your use of the word “self-righteous” is incorrect when you use it to define an assertion of fact. It is simply a fact that sales reps are required to report AEs. Nobody’s being self-righteous about it. What you want is for someone to be punished for your son’s death–which may or may not be attributable to Zyprexa. Similarly, you assign causality to Lamictal regarding your daughter’s state of mind–this may or may not be true. None of us can know that, since we don’t have access to her medical records.
All chemicals taken in by or put on to the human body have side effects–whether those chemicals be protein from food, compounds used in drugs, vitamins or whatever. Sometimes the side effects are bad, but we continue to use the product because we decide it is worth the risk. Such assessment happens on an individual level (I won’t eat hamburger because I am paranoid about prions) and on a corporate level (we require prescriptions for some drugs; some drugs are no-nos completely, i.e., heroin). The corporate decisions are made based on the good of the whole, not the risk to a single individual. Thus, while a serious adverse event from a drug is a tragedy for the individual, the group may well decide that it is willing to pay that price in order to have the drug available to all.
I think the “self-righteousness” you decry from some of us on the site is a frustration with an attitude that seems to be: something bad happened to me; I think it is because of Drug X, pull that drug from the market and punish everyone involved. Everything is fraud; everything is a conspiracy; no one involved with pharma is sincere or genuine.
The truth is that most times, causality and fault are not easy to determine. Certain cases make the headlines, but that doesn’t mean every instance is a case of fraud or malignancy. Physicians have responsibility for their prescribing decisions–no honest psychiatrist could ever say that he/she didn’t know that Zyprexa causes weight gain, no matter what the reps were telling them. Patients also have resposibility–to ask questions, to report side effect EARLY, to use drugs correctly.
The sheet of paper has more than one side–every single drug company employee is also a patient, a consumer, a family member. Most Americans are actually the owners of the pharmaceutical companies–as shareholders. It is as Jaynesday often points out: we need to come up with constructive paths forward.
HorusCat
Justice,
I would say that the instances you describe are the exception, not the rule. Not everything is a conspiracy.
If an AE is frequent, it is impossible to cover up. Like I said, everyone knew Zyprexa causes weight gain.
Rare, serious AEs are by their nature difficult to tease out. I think most companies do their best to discern them and report them. I sell a drug that just got new labelling when I got a new indication–and the labelling changed to reflect a rare but serious adverse event. The drug is a huge money-maker for the company, but there was no cover-up, no “cat and mouse.”
Bob Freeman
Justice, causality is so difficult to establish that it almost doesn’t matter who does the analysis. Case reports and case-series analyses aren’t adequate to establish causality and you can’t calculate incidence rates or relative risk. Those quasi-designs are suited; however, for studying rate and/or latent events.
The FDA will allow a very, very crude estimate of incidence rates by allowing the % of the US population taking the drug to approximate the denominator. Crude is probably an understatement.
You need either prospective cohort studies, controlled for as many confounding variables as practicable, or randomized clinical trials. to get at causality. Even then it’s difficult.
Justice in Michigan
Hi Bob - As I understand the Medwatch system, causality is explicitly exclused as a criterion for reporting. In short, it is beside the point. Rather, docs and/or others are asked to report comcommitant conditions that emerged regardless of speculations about causality.
I agree that such data are always limited compared with more controlled studies. Still, they offer suggestions about possibilities. For example, when higher rates of CV events among those taking Vioxx in observational studies, the top Merck scientists - Skolnick on down - concluded that this was _probably_ mechanism based, as compared with either statistical anomaly or some other explanation. Nonetheless, Merck’s promo materials did everything they could to deny that possibility.
Understanding the limits of various kinds of studies, the bottom-line question comes down to whether companies go out of their way to prove _potential_ causality or go out of their way to deny or obfuscate it.
The “interesting” cases are when a company does its own observational studies that suggest possible causality and then - when the possibility is raised by others - they trash such studies as having no significance.
There have been numerous such instances. So what I’m asking folks with industry experience is whether that is, from your/their perspective, an OK way to do business.
Justice in Michigan
HC - Sorry I didn’t include your comments; I didn’t see them when I first responded.
One of the interesting bits in the Baycol case that was included on this web was a description of Bayer’s then CEO saying, in essence, that is was OK to create promo material that denied what the company already knew about Baycol. That’s what “everyone else does” so, unfortunately, “we have no choice.”
I genuinely hope you are right that such instances, despite the CEO’s own statement, are anomalous. I have no fondness for conspiracies; only truth.
Based on other sources, my sense is that “the field” often does not know what is known internally.
Bob Freeman
It’s the age-old problem of the desire to equate statistical association with causation. While statistical correlation is necessary it’s not a necessary and sufficient condition for causalilty; that comes from the research or quasi-research design used to control for confounding variables and to ensure internal validity.
As has been noted, MedWatch is voluntary for physicians and pharmacists and mandatory for manufacturers. You are correct in pointing out that the responsbility for follow-up is often relegated to the manufacturer due to backlogs and lack of resources at the FDA. I don’t believe the assumption of causality is explict; it’s more of an perceived association.
As I have noted elsewhere, the ’spin meisters” within companies have rushed to defend products prematurely and aggressively before adequate studies have been undertaken. Another change in corporate behavior which I bemoan.
Justice in Michigan
Bob - My understanding (you or others correct me if I’m wrong) is that manufacturers are obliged to report significant AEs independent of _any_ presuppostions/hypotheses/speculations about causality.
That said, I believe there are instances in which companies have opted out of forwarding AE data precisely because of causality assumptions.
Perhaps there is some ambiguity in the law. We do know that the Bush FDA is working to obviate companies responsibilities to report potential problems _unless_ there is reasonable evidence of causality.
To me, this is clearly an attempt to dilute the CBE part of FDCA, and thus part of the attempt to lock up FDA preemption. The criteria for unilateral warning is set beyond what any company could normally meet, and thus FDA hegemony becomes the default position.
As my Dad would have said (Father’s Day reference), “it’s a racket.”
Bob Freeman
The SAEs that must be reported are the ones resulting in 1). death, 2) a hospitalization or one that prolongs and existing hospital stay, 3) one that results in permanent disability, and 4) one that requires medical intertervention to resolve or reverse. There may be another criterion but it’s late and my memory fails me. But, yes, SAEs have to be reported regardless of any hypotheses about association, causality, etc.
As far as non-SAEs I’m not sure what the regs require. If they are not required, then I imagine reporting will vary by company. The late, great Upjohn Company was very deligent in reporting everything. Others, I just don’t know.
Justice in Michigan
Thanks, Bob, even at the late hour.
Assuming what you say is accurate, it sounds like it allows a fair amount of discretion in terms of reporting.
In any event, the question I originally raised was whether the “cat-and-mouse” model seemed appropriate as business practice. These are fairly specific questions:
1. If a company duly reports SAEs but withholds its already done analysis of their potential significance (on the assumption that this is FDA’s responsibility), is that OK?
2. If a company knowingly opts out of analyzing SAE data (submitted) based on the anticipation that the result will not serve them well, and it is better that to let the chips fall and “save the drug” as long as possible.
HorusCat
Justice,
Of course it’s not ok for a company to obfuscate (what a great word) or cover-up AEs they suspect are related to their product. Nor is it proper to “spin” against such a suspicion. And you are right when you say the field often doesn’t know–especially with these rare SAEs. I think some companies are better about it than others.
I think part of the problem is when a product is getting a lot of press about potential AEs (i.e., Chantix), but no one is really sure what is going on. At that point, what DO you tell the field to say? I don’t know what Pfizer has told their reps, but I suspect it is along the lines of having the doctors call medical information. The rationale behind that would be you can’t have 1,000 reps out all saying different things, and perhaps incurring liability by what they say.
Bob Freeman
Hi, Justice, the first step is to report the SE via MedWatch, then investigate. Obviously it takes time for these isolated events, “signals” if you will, to be analyzed. The FDA can, as I understand it, ask the company to undertake an analysis, or the company can start independently from the FDA. Per HC’s comments elsewhere, reports of minor events, say dizziness, are not going to get attention compared to a SAE.
To question 1, failure to submit an analysis prepared by the company would be unethical at least. (I’m still referring to SAEs) An confounding issue is what conclusion, if any, the drug safety people (e. g., epidemiologists) have reached, and knowning these people quite well, they’re not going to issue a report until they’re satisfied that a relationship does or does not exist. As noted, it is very difficult to get much past statistical association/correlation at this stage.
I’m not sure I understand Question 2. Could you clarify.
To HC, I’m seeing a lot more spin coming out of companies on potential SAEs than before. Some of the content is legit: “It’s not known if the SAE is related, etc.”
Melody
I’ve acknowledged I’m a layperson, and perhaps my take on adverse event reporting is naïve. The real problem, from my perspective, is two-fold. The reporting process itself, is laborious (for medical professionals), entailing time-consuming follow-up. At the same time, individuals who report can be trivialized as anecdotally reporting things beyond their understanding. But the biggest problem, IMO, is the way the FDA collects, categorizes and “analyzes” the information. A friend sent my husband a collection of one years’ data on AEs arising from the use of insulin. As we studied the collected data, we were astounded at the numbers and ways ‘categories’ were used to deflect, diminish, and disregard significant information.
In my simplistic view, I would offer this analogy. Let’s suppose that 10 diabetics were switched to a new insulin or insulin-like product and experienced an anaphylactic reaction at first injection, to a degree requiring emergency medical intervention. If all 10 events were reported, categorized, and analyzed under the label ‘anaphylaxis’, one would suspect alarm bells would be ringing within the FDA, demanding further analysis and response. But . . . suppose the AE reports were instead revealed thus:
2 difficulty breathing
1 labored breathing
1 cardiovascular event
2 tingling/loss of feeling in extremities
1 confusion
1 disorientation
1 memory loss
1 allergic response/rash
In the first case . . . 10 reports labeled ‘anaphylaxis’ –pointing out SIMILARITIES—among all 10 reports would be significant. However, in the latter example, 1 or 2 reports (of any description) out of 10 reported—signifying DIFFERENCES– is certainly less noteworthy. The use and abuse of language itself may be indicative of just how dysfunctional our current FDA AE system is.
laura
So Bob, once a company has made a report, are they then obligated to update the report if the outcome changes? In my mother’s case, after suffering her reaction to the medical device, she remained in the hospital on life support for three weeks before passing away. It was during that time that the company made their report and submitted it to the FDA. Therefore, her outcome, as far as the company was concerned, was “required intervention”. They never went back and amended the report or filed an additional report with an outcome of “death. Is this possibly a way that companies “manipulate” the system. Obviously an
outcome of “required intervention” is much less significant than an outcome of “death” and if they met their obligation by filing the report, the company would still be in compliance.
Bob Freeman
Laura, I can see a number of logistical problems which are a concern. For instance, for the company to submit the SAE, it would have to be aware of the case and its outcome by being informed of it by the physicians attending to your mother. Apparently the company was notified of the SAE early on but not informed of the outcome. Why? I have no idea.
Privacy would prevent the company from dealing with you or your mother directly. They could not access her hospital records. One would think (or hope) that the attending physician would report the outcome to the FDA as well as to the company. It is, as you note, a very imperfect system. The information flow was flawed.
Bob Freeman
Melody, for a layperson as you describe yourself you’re certainly an enlightened one (I’m not patronizing you). Categorization of raw data into discrete variables like these is one of the challenges of research–coding, while standardized, has some subjectivity specific to the coder, and inter-rater reliability needs to be addressed, usually by elementary statitical tests to see if people are rating the events the same way. I have no idea how the FDA handles inter-rater reliability.
The more you breakdown data into discrete variables the noiser and more equivocable the analysis becomes. Instead of having few variables (shock, e. g.) upon which the correlations are calculated you have many. There are statistical ways to handle this and reduce the number of discrete variables, but again, I don’t know the specific ways the FDA handles it.
Justice in Michigan
Melody’s example is a very informative one. Once again, we learn that “reporting” SAEs - and being technically “in compliance” - can mean a lot of different things in practice.
Dare I say (well, why not) that this is yet another reason why FDA preemption is “perverse” (New York Times description) policy. None of these sort of issues (even if no culpability was involved) would be likely to see the light without the discovery process and/or years of FOI requests, probably resulting in not much until many years had gone by (and whatever bad happened had happened to many).
Bob - you’re right; my second example got lost in the late night typing lack of clarity. The gist is a company that is _considering_ doing its own internal analysis of a suspected problem, but opting out of it because “if FDA asks, we have to give.” This differs from the first example because, in the first case, the internal study had already been done, but kept inside the company and/or edited in such a way as to suggest the company had fewer concerns than it actually did. (There are no shortage of such examples.)
Either way, we come back to the questions:
1. What did they know?
2. When did they know it?
3. What did they do (or not do) about it?
The issues of doing internal studies but not reporting them, or not doing them at all, certainly do not constitute felony fraud. Once again, such companies would be “in compliance” with FDA. And, once again, it would the cat/mouse game of the FDA figuring out what the company already suspects or - from their own perspective - already knows.
So once again I ask - especially Atlex, Nathan, and others in industry - whether they that way of doing business is OK from their perspective?
And I ask them and others - how far _is_ a company obligated to go, unilaterally, if they anticipate that going far may, indeed, sink their product?
Such questions are nbot without some complexity. But no obfuscation (thanks HC!). At least I didn’t say “bloviating…”
Bob Freeman
Good questions, Justice. Let me try to frame a brief response to each and see where it takes us.
1. What did they know? One doesn’t know until sufficient analyses have been done to 1) suggest that an association exists and 2) undertake the kind of research that will provide with varying degrees of probability what we know. If one wants causality, it’s going to take a long time because you really know very little from case-series analyses and retrospective cohort studies from large billing data, other than hypothesis generation.
2). When did they know it? Two steps: one time frame is from the preliminary analysis and one dated by the time for subsequent analyses. Second, after the intital work is reviewed and validated by insiders and extramural reviewers. (sorry for the incomplete sentence.)
3). If there’s association a discussion with the FDA should follow. At this point, I refer back to other posts on this and other threads as to how effectively this process works or doesn’t work.
You’re correct in noting this is extraordinarily complex.
Justice in Michigan
Thanks, Bob. Clearly, there are “dramatic” cases in which it was not too complex, as we end up learning later on.
But let’s look at it another way. Do you know of instances in which a company proactively investigated suspected risks an an _already approved_ drug (not waiting for FDA) knowing that there was a real chance the results could “kill” the product?
As I read the latest on the Vioxx cases, it sounds like there were some in Merck who genuinely wanted to do this (and within a reasonable time frame). Likewise, there were some in Bayer who did so during the Baycol era. But, as far as I know, they were overruled or stalled as long as possible.
Also aware of current long-term study on Celebrex, but that is, indeed, very long-term and obviously arose in the context of the all the cox2 issues in general.
Certainly, the purely financial incentives go against initiating such studies.
In the meantime, while I would take a different view myself, certainly there are some who argue some version of:
“The only obligation the company has is to meet FDA requirements re: SAE submissions. Beyond that, it is entirely the job of the FDA to determine whether is, or is not, a problem.”
Melody
To make the problem even more complex, the roles played (or not) by doctors and patients impact the system. Doctors may not want to report AEs because of the ensuing ‘reporting’ burden . . . or potential for litigation.
Patients–often undereducated–fail to follow through just because of the ‘human factor.’ When a bad drug reaction . . . or bad medical practice (combining drugs, off-label). . . or just an unalterable disease outcome results in death, the survivor experiences his/her own very significant trauma. By the time he/she sorts through and addresses surrounding his/her own tragedy, the resolve to follow-through is usually non-existent. So statistical information–correctly identified and submitted for analysis–goes missing. It is another piece of the puzzle that is needed. How can FDA correctly assemble the puzzle, when many pieces are missing?
Justice in Michigan
No question, Melody. The Medwatch “problems” have been bemoaned for many years.
Yet again, the FDA has suggested more active surveillance than Medwatch provides, using a range of data-bases, domestic and beyond, independent studies, etc.. I guess we’ll see.
Approaches to monitoring safety that do not rely on Medwatch have been proposed for a long time. As I understand surveillance in some of the European countries (not very well), it is significantly more reliable.
Jaynesday
Databases are means to “digitize” (make digital images of) complicated information.
Working with databases can be very frustrating. Not only for the designer but also for the one assigned to enter the data. My experience is that there is a fine line between a useful database and one that is too burdensome to use. However I don’t know of any other way to provide guidance from massive amounts of data. If we’re talking about numerous databases it would seem the problem would be even more difficult.
Is there some other way to make decisions from massive amounts of information? Could we come up with a simplified initial data set that would only be used to flag certain data that would then lead us to focus in on information that needs to be followed up on? This would require more that one reporting situation. All data would not need to be reported in the first report but coming back for questioning would be required.
Jaynesday
One of the first questions might be. Would you be willing to be contacted again about the AE if further information is required?
CMC guy
JIM it may sound like obfuscation in answer to your embedded questions to industry insiders (since Bob well covers others) but because of the complexities there are often not easy answers regarding acceptable business practices/obligations. I think in general industry has high standards that should provide proper reporting and actions regardless of a negative impact do exist. Although exceptions happens all around in implementing most Pharma has firm Policies and Guidelines aimed not only at maintaining compliance but also assuring patient protections relative to SAE & AEs (others have rightly pointed out weakness in the system). Likewise handling and evaluation of data is frequently strictly codified along with FDA interactions. Problems occur because people who are responsible do not hold to established standards or otherwise make mistakes. My own believe this is mostly unintentional errors due to complex nature (can’t see single tree in a forest) although acknowledge has been way too much intentional bad acts (cut the tree down so not seen). I also feel can and should do better.
Anyone who is stupid enough to attempt to play cat & mouse with FDA deserves and ultimately majority think do get caught, so experienced people don’t play games. FDA typically wants to do their own analyses of data so often summary reports, if included, only focus on obvious and do not delve into more speculation (which internally tends to go round & round anyway). Hindsight can show could have handled differently but again clear cut signs are rare so items treated as generalized unexplained events and not hidden. I work on development side and so major focus is on potential problems, along with benefits, so continual risk/benefit analysis mode and most places approach is to kill off things rapidly (to save money). Attitude does change post-approval, because does mean loss of revenue, however in main companies will do the right thing (but can be slow because have to convince themselves is real).
Bob Freeman
Justice, I’m tied up for the next few hours but to answer one of your questions, yes, one company for which I worked aggressively pursued studying a side effect of a marketed drug without any prompting from any external source, FDA or otherwise. Their reporting got to the point that the FDA told the company that they did not see the side effect as being of concern and suggested the company was pursuing a dead end. I don’t feel comfortable mentioning either the company or the product, other than to state the drug has been off-patent for years. Also, it was not a SAE in the regulatory definition of the term.
Jaynesday, your comments about data bases are right on target plus the data quality has to be verified and often require “cleaning up” (not a sinister motivation–data are known to be “dirty”, missing, mis-coded, etc.)
Bob Freeman
Excellent point, CMC Guy, if there are SAEs you want to know them early in the developmental process for the very reason you stated.
CMC guy
JIM- I too know several examples where companies either proactively pulled something off market or initiated studies because of unusual observation without waiting for FDA to act (Don’t wish to get you started but in one case was because of perceived litigation potential so wonder if would happen if preemption granted).
The view stated regarding obligations is not one I have heard. It sounds like a lawyer talking though and you may recall my distaste for them.
Justice in Michigan
Thanks to CMC guy and Bob for very interesting and useful comments. Don’t worry about “getting me started,” btw, since I’m rarely stopped. I know of instances of the sort you describe in any event, but it is always hard for we outsiders to understand them in the broader context.
Again thanks (from this not-lawyer).
CMC guy
JIM it is hard to deal with in a specific context as can have TLI (too little information) or TMI (too much information) both of which can lead to confusion and paralysis. I must add I have never been directly responsible for making a final decision of this type, only in evaluations/studies/recommendations but it can be very difficult to understand, especially because it usually does come back to money issues. It’s usually not black & white and dealings are in gray areas.
To be consistent in my earlier comments I should have mentioned a rumor/lore at one place where an “unnamed group” wished to bury some potentially negative info in a “data dump” report in spite of clinical opinion that it needed to be more forthcoming. When Reg VP got involved he used some terms similar to SP/Merck Vytorin exchanges and made sure everything was above board. Illustrates the good and bad.
Justice in Michigan
Great example, CMC. What you describe fits with my outsider sense of the various “factions” within many companies, and the texture of decision-making - much more complex than we outsiders generally realize.
These are stories usually entirely missed between the trashing and the rigidly defensive approaches to discussions of pharma. One day, I hope to write about them.
Bob Freeman
A lot of the complexity, Justice, in big pharma is the complex, cross-functional matrices at the product and drug candidate levels. Even sitting on one of those things, it’s hard to figure who’s responsible. That’s an overgeneralization, I admit. You’ll see a lot of duplication of effort (and competition) by who has control over a budget.
Just A Thought
If I have any of this wrong I would truly like to understand the process better.
Reporting myself came from my doc being less aware of the situation than I was. It’s not his fault, no one was aware.
Also, while speaking with him, and because he was hurried, I noticed he was taking note of some issues and not others. He did say he’d speak to the drug co/rep. Okay, cool, but is he giving half of the story? That is dilution #1.
Then the drug rep delivers what remains of the information to the company. Now, I do not know the procedure there, but maybe he’s finding that some of my diluted issues are less significant than they really are. Dilution #2.
By the the criteria for establishment, shown above by Jack 2, things are seemingly simplified, as I expected.
Then the company, who is of course going to do what they can to support the product they produced, may look at this twice diluted information and think it is nothing at all, or they apply what they have prepared to explain things away, causing dilution #3.
By the time the FDA gets my information, if at all, it no longer resembles the severity of the problem.
So yep, I reported directly to the company and to the FDA. My doctor went about reporting the things he found important.
Though it really didn’t matter anyway. Who am I to know what has gone on with me? Am I a doctor? Nevermind that nothing has changed but the drug I was taking. The reporting system is seemingly pointless. So then you go over their heads, get noisy, and demand attention. What else can you do?
HorusCat
Just A Thought and Melody,
I think many of the problems with the system come from our rigid protection of patient information. Dilution does occur, because the FDA doesn’t get patient identifiers. Neither do drug companies–if a doctor tells me about a side effect, he can’t tell me confidential information. So basically I find out age, sex, type of AE, other meds, length of time on my med, dosing. That’s if the doc can remember all that. Medical information questions me very closely about any information I have been able to gather. But there is no follow-up, because patient identifiers can’t be used. Thus, for example, the company involved in Laura’s mom’s case may never have known the outcome–or it may have been reported as a separate event. There is no way to inform the FDA or the company, oh by the way, that “required intervention” report turned into a “death” report.
And what may seem like deliberate disingenuousness in reporting vocabulary actually probably goes back instead to being exactly what a physician says about a problem. If a report is called in as “anaphylaxis,” it gets taken down that way. But neither the FDA nor the company, if it happens to get the report, can make diagnoses based on the information they get. So they just report verbatim what the patient or doctor says: “I had trouble breathing.” “I got hives.” “My chest felt heavy.” The company is not allowed to turn around and call the patient’s doctor and question him further.
I think European reporting is much cleaner and more useful, precisely because they do not have HIPAA.
Jane
If there is no “patient Identifers” information being provided, how in the world are the people reporting psychiatric side effects from Chantix being labled as having pre-existing mental disorders? Are we to “assume” the drug maker just made this all up to explain away the side effects and to account for all the reports?
Perks
Jane - don’t ask! You’re inviting more damage limitation ‘doh’ responses from those with conflicts of interest. People whose great living standards rely on pharma dollars regardless of the cost to others.
HorusCat
Perks,
Once again you prove your ignorance. One needn’t know a patient’s name to speculate that said patient has this or that disorder. In other words, I could say, “Those people reporting depression are probably already depressed.” Or “trying to quit smoking causes depression.” Give me an example of a Pfizer person identifying a complainant by name and saying that person has psychiatric issues to begin with. You can’t, because it hasn’t happened.
Furthermore, some of the very outspoken patient critics of Chantix have self-identified and revealed personal information. Thus, they compromise their own confidentiality. In the case of a suicide that makes the news, the patient’s confidentiality is broken by the family members who complain. Pfizer has never been the initial reporter of the name or personal details of a patient on Chantix, because Pfizer doesn’t have them.
See, the problem with your sarcasm is that it is way too easy to get the answer to Jane’s question without involving a pharma spokesperson:
Get on the FDA website and look at the AERS. There are no identifiers. Call the FDA and ask if you can have patient identifiers–ask if they even have patient identifiers. Ask your doctor if he gives out patient identifiers to reps when he/she tells them about AEs. Call a company medical affairs line and ask if they require patient identifiers when a patient calls to report side effects. Heck, call a company MA line and report a fake side effect–see if they ask you for your name and medical history.
Meanwhile, Perks, I hope you own no shares in mutual funds or pension plans, because if you do, you are one of those “conflicted” people. And I know that when you get cancer or dementia or hypertension or diabetes, you will refuse to be treated with medication, because you don’t want to line the pockets of people like me.
HorusCat
Just A Thought,
Do I remember correctly that you take an anti-convulsant–maybe Dilantin–and had problems with a switch? I’ve been asking my docs if they are seeing problems with the new Dilantin…none of them have had any problems. Doesn’t mean it hasn’t happened–part of the difficulty with CNS meds is that each patient is so exquisitely unique and sensitive.
HC
HorusCat
And of course, Perks, feel free to ignore the facts that I post, since they don’t fit your presuppositions, and instead indulge in ad hominem attacks about someone else’s standard of living, etc. That tactic lends you such credibility and advances the discussion so greatly that we would hate to lose your contributions.
Bob Freeman
HorusCat, I need to qualify what you said about Europe being more effective in their AE reporting. Most European-based pharma people I deal with state the US HIPPA regs don’t go far enough. Speaking from personal experience both primary data collection and data mining of national health system or Krakenkassen records are very difficult for patient confidentiality reasons.
Justice in Michigan
Bob - Can you clarify for us non-native speakers your comment: “A lot of the complexity, Justice, in big pharma is the complex, cross-functional matrices at the product and drug candidate levels.”
Is this related to the often discussed tensions between the “science” and “marketing” sides, or something else (and, it sounds) more specific?
Grazzi!
Bob Freeman
Good morning, Justice. I’d like others to weight in because operating models are somewhat specific to the company. My answer is going to ramble a bit but here goes:
At the brand level you have a senior management team that has some if not all of the following types of representation. Generally the team is lead by a product director (or higher level director title, perhaps even a VP). Some teams are co-led by a Marketing Exex and a Medical Director. It is populated by a Regulatory leader, Manufacturing, an attorney, a Medical Affairs leader (Phase IV), a Finance Lead, perhaps a PR Lead, etc. Now, think of the core leadership team as the hub of a wheel.
The spokes that are connected to the hub are other product teams, more operational than specific. The same types of folks will be on these teams. How many teams will depend on the company–a reimbursement team will have a pricing strategist, another finance lead, product manager(s) (sometimes several product managers are there by indication–For a SSRI you might have a product manager for Depression, one for Bi-Polar, e. g.). These teams will carry out the tactical plans approved by the core team.
Now, above the Product Team you often find a Therapeutic Area Team (oncology, CNS, Cardiovascular, Women’s Health, Metabolic). You will have a matrix team in effect that is responsible not only for the branded (marketed) products but products in the developmental portfolio. This is usually a global team although there can be a US team at the therapeutic level that reports in to a global team.
Now to confuse things more, the people who populate teams are deployed from functional and support division and these folks have a direct reporting line to VPs that head these divisions.
Thanks to the consulting firms that really run pharma companies these operating models get taken apart and put back together every few years (when the consulting firms want to milk more $$$ from the company).
Confused? All of us are trying to find out where the real decision-making is.
Justice, you usually don’t see the conflict between medical and marketing at these kinds of meetings. The subterfuge and politics are played out at the divisional levels: duplication of effort, competition among divisions for budget and influence, etc.
Yes, the Divisional VP, global and or US, have their own matrix organizations to make portfolio and brand management decisions.
Imagine Corporate as a holding company that tries to coordinate the work of a number of small to mid-size pharma companies.
CMC guy
Bob as per usual your comments are spot on. I have been a number of places and Hierarchal Team structure existed in one form or another with the elements you mention. In most cases it does come down to a single or small group of people who really hold the power but knowing who they is not always obvious unless one knows the real poltics of situation. They are also subject to change, particularly during CEO or others replacements at top levels. Smaller companies commonly are not as entrenched but also operate in an analogous manner.
Great comments about “consultants”. I can’t count the number of management approaches, paradigm shifts and other new (recycled typically) approaches I have been through. Fortunately most people can focus on doing their jobs, attempting to make progress on projects, in spite of all the clutter and politics around them.
HorusCat
Bob,
I yield to your superior knowledge. I was thinking in particular of pregnancy registries–in Europe, as I understand it, physicians are allowed to report pregnancies and outcomes. This leads to much more useful information than here, where a physician is prohibited from reporting. I think we could probably devise a system where patient identifiers are obscured and yet the patient can be tracked throughout the course of an AE or a pregnancy. This would require some software geniuses (Bill Gates, where are you?), a uniform system, and motivated participants–including physicians and patients. Then, from the sounds of it, a plethora of statisticians and analysts to sort through all the noise and determine what signals, if any, are being generated!
Laurie
“Give me an example of a Pfizer person identifying a complainant by name and saying that person has psychiatric issues to begin with. ”
While Pfizer may not being doing this publically, there are a few on pharmalot(who admit to working in pharma) who have used this as the explanation of suicidality with Chantix repeatedly.
HorusCat
Laurie,
So, Pfizer isn’t doing what you are accusing them of. There are anonymous posters on this site who make unsubstantiated claims, but they don’t provide names or medical histories, either, unless that information has already been made public. Basically, the only time we have that information is when the patient or family has provided it.
Just wanted to be clear about that.
Jaynesday
Bob, I can see where the political pressure in such an extensive team would be huge. One aspect of a drug’s development could sink the whole project. After years of development and massive costs, I wouldn’t want to be the one that has to says…”we can’t go any further with this product because…”
I can see the glaring eyes now. Usually, and I’m talking in a general sense now, people find ways to ignore such conclusions.
Is that where the consultants come in?
Jaynesday
HC, I like your thought….
“I think we could probably devise a system where patient identifiers are obscured and yet the patient can be tracked throughout the course of an AE or a pregnancy. This would require some software geniuses (Bill Gates, where are you?), a uniform system, and motivated participants–including physicians and patients. Then, from the sounds of it, a plethora of statisticians and analysts to sort through all the noise and determine what signals, if any, are being generated!”
I don’t think it would require too much of a genius to come up with a secure system. We do it in on line banking now for example.
I think we might have to scrap the current system and start over to take advantage of more powerful software and hardware. A database designed in 1989(?) had to be limited by the computing power and memory available then. Also back then it was almost unthinkable that a computer network could be secure. Also simplifying by using a series of reporting stages would help.
Concerning the AE “noise”, I think that if the public was insured that their AE’s were taken seriously, that they might even be asked to clearify or add information when they have submitted an AE; the noise would be drastically reduced. People would not be so apt to “vent” into such a system.
Justice in Michigan
Bob - Thanks so much for this overview of the structure. All large organizations are complex, of course, but this really helps me better understand the relevant layers and players.
Comments re: decision-making and power by CMC also of interest.
And Jaynesday re: political presure. Not to focus on any single instance, but one of the better known Vioxxiana trivia is Alison Reicin’s comment that is almost a direct quote of your scenario - “I don’t want to be the one who tells …” etc.
From a sociological perspective (where I know a little more), it is, of course, the complexity of such organizations - and some of the “guesswork”/unpredictability about politics/power - that works to diffuse a sense of personal responsibility (beyond one’s immediate and official tasks) and keep people focused at their benches.
Laurie
I never said that Pfizer was doing it. I was just trying to clarify, from earlier postings and media coverage on Chantix, that the rationalization of the side effects of Chantix are out there, regardless of the source.
That rationalization can prevent others from coming forward to report their own situation with side effects.
Once again…a system that is poorly designed to track AE’s, especially if the reaction falls into the “psychiatric event” category.
There has to be a better way..what that is I still don’t know.
HorusCat
Laurie,
I understand. It’s just that your comment was a little sarcastic and inflammatory. And we don’t know if it is rationalization of Chantix’s side effects or not–that’s kind of the point of this whole thread.
Jaynesday,
You are absolutely correct about the software/reporting problem being manageable. But people get much weirder about their medical data than they do about their money! RevolutionHealth is working with Microsoft, I think, to develop universal tracking software. Perhaps someone should put a bug in their ear (which, BTW, I have heard is VERY painful–there was a segment on Oprah or somewhere) about a function for tracking AEs.
Well, I have to go push my body around the block a few times in my relentless quest to die of something other than heart disease. Have really enjoyed this thread and will check back in soon….Kind of bummed. Tonight is movie night and there’s not a whole lot out there to see. But I am really in the mood for hot popcorn, and it just doesn’t taste the same at home, no matter what my husband says.
HorusCat
Laurie,
Real quick: I think the defensiveness of the companies correlates to the strident calls for products to be pulled from the market before the evidence is in. We need to all slow down, be cautious and prudent in our use of a questioned drug, and give ourselves time to assimilate the data. We might be able to keep drugs on the market (Bextra, anyone?) while limiting any actual damage done by indiscreet use. I would say Chantix is such a situation. Let’s stop with the rhetoric, come up with a monitoring system, examine the data and find a reasonable way to use this drug in an informed fashion. There is no doubt that it has helped some stop smoking who could not stop any other way, and this has positive individual and public benefits. There are probably individuals who have been harmed by the drug. There is a gray area in between uncontrolled prescribing and pulling the drug from the market: let’s try to find it. If we don’t find this gray area our pharmaceuticals, we are going to find ourselves with no new medications. The investment won’t be worth the risk for the companies.
Laurie
“the strident calls for products to be pulled from the market before the evidence is in.”
I agree that the research has to be done. But we have also seen in the past the years it can take to confirm a potential adverse reaction. If there is informaton that their may be a problem, that needs to be relayed to the physican prescribing AND the patient being given that prescription. The first ssri adverse reports were brought to the FDA in 1991 and dismissed….2004 they were confirmed and finally acted on, after many deaths and horrible reactions.This after what I’ve read were record setting numbers of adverse reaction reports for a single category of drugs. But it took 13 years to warn? Somethings wrong there.
I think that’s where my concern is. We know the current system is inadequate to properly monitor adverse reactions, but even when those adverse reactions are officially presented face to face with physican documented evidence, or in face to face reporting to the FDA they are still discounted.
This isn’t as simple as getting a good reporting system. It’s about what is done with that information and how the FDA acts on that information, two separate things.
I’m all about education for the patient. If a patient is made aware of the risks AND benefits of a drug and chooses to take that drug, that’s their choice. If a patient takes a drug only based on benefit reports, then we’ve put that patient at unnecessary risk.
The risk of suicide may be reportedly small with Chantix, but if it’s your family member who commits suicide, then that risk is huge. Warning all, even though the risk may be potentially small, is the only logicial thing to do, considering the lack of predisposing factors. And by officially warning all, honest reporting of adverse reactions carries no stigma.
HorusCat
Laurie,
I don’t disagree with you. But how do you word a warning when you aren’t sure a link exists and the media don’t take the time to flesh out all the details and nuances of what is happening?
Bob Freeman
Justice, Jaynesday, HC, et al.,
One other comment, Justice, that makes things somewhat more complex is the constant turnover of key leaders, who are being fast-tracked. I don’t know what the average tenure is anymore, but it used to be less than a year. This make continuity difficult because each functional leader wants to put his/her imprint on the product/therapeutic level. There was less turnover at the tactical team(s) level, but they were constantly re-educating newly appointed leaders.
HC, I wasn’t aware of the pregnancy registers. Thanks for pointing that out.
Janesday, yes, dissent, especially those perceived as killing off a developmental compound, was never welcomed by those whose projects were targeted.
Jaynesday
Bob, Thanks for shedding light on product development and management in pharmaceutical companies. Very interesting.
Hope I’m not being over dramatic when I say, what a recipe for disaster.
*High cost/profit stakes
*Complicated product teams
*Managers pressured to shine and/or high amount of management movement
*Product testing based on statistical analysis (Ability to skew results)
*Products that can either be immensely helpful and/or cause great harm to the customer.
*Human nature
Not Directed at Bob – (Just my rant)
I have to say again that with such a highly critical (to human life) product combined with the complexity of the development process; there has to be a means to generate internal pressure for not only successful design and manufacturing but immediate and lasting customer satisfaction. Transparency to the customer can accomplish this. If the customer is allowed to know how a company rates concerning performance of their products everything else will take care of itself.
A truly successful company must *want to* be known as the company that produces the best products with the absolutely *least* amount of harm to the customer (even above statistical probability) and only those companies should be allowed to gain the financial reward that inspires such an attitude.
I for one don’t think we’re there yet. The attitude in some companies (sorry this is an outsider’s view) seems to be what can we get away with in the short term while making huge profit in hopes of covering product liability claims in the future?
I would call this “Drive by Drugs”.
Chris
Pharma companies almost always know how ‘customers’ rate the performance of their products and the company overall (and competitors)- that is the function of the market research group. However, it is considered the responsibility of the marketing and sales groups to positively influence those perceptions. In the pharma world, it is perception that drives prescribing, not the realities. If transparency to the customer negatively impacts product perceptions vis a vis other competitive options, then sales and market share are negative affected. This is why you have downplaying of adverse events and overpromotion of benefits. Even though it may be appreciated, you don’t gain market share because of honesty.
Jaynesday
Thanks for the explanation Chris. I was thinking more of awareness by the customer of all of the feedback from other customers in the form of a rating system. Kind of like Consumer Reports for Drugs and Devices. Something that is available at the time of prescription that can be a guide to the patient. Should I take this drug that has a poor rating or take that drug that is rated better, even if it cost me more? If this were possible there would be a naturally generated pressure internal to the drug manufacturer to increase the performance and safety of their products.
Bob Freeman
Jaynesday, thank you for a very thought-provoking reply to my post.
I will get back to you later today with a detailed reply–I have some obligations that will take up much of the day.
The teams have very complex processes to facilitate decision-making and ensure milestones are met. While many of these processes work well, others (to me) are process for the sake of process; i. e., checking off steps in the process gives the false sense that the work plan is being successfully completed. Take adverse events reporting, for instance: the folks in drug safety have process steps to follow when an AE comes in from either the field, a physician, pharmacist or customer. They use, if you will, a cookbook approach to ensure the AEs are followed-up and the analysis is reported within the company and eventually to the FDA. I would argue that most of the time the process works but human variability, intentional and unintentional, means problems can occur.
Re, “corporate branding” to the public: companies’ efforts are largely underfunded and inconsistent. Controversy exists in every company as to whether the company’s name should be associated with the brand–a lot of marketing research people argue that’s too much info for consumers to process and remember in a TV spot, for example. Oddly, using the company’s name in a drug ad is a decision made at the brand team level because the team is the one that pays for the ad, and some teams believe the mention of the company’s name distracts the viewer from the product message. (This is an incomplete response and I will come back later to add detail and qualifications.)
Laurie
“But how do you word a warning when you aren’t sure a link exists and the media don’t take the time to flesh out all the details and nuances of what is happening?”
Your have the doctor explain to the patient that there have been reports of a,b,c that are not definitively confirmed, but should be watched for when they are taking this drug. You encourage them to report those side effects to the doctor or the FDA.
Some of the side effects that ultimately end up with a black box warning are so bizarre that most patients wouldn’t associate that reaction with a specific drug.
At least if they are made aware of it, they don’t ignore a potential fatal reaction. ie. ssri’s and suicidality. Easily dismissed as “part of the disease”, when in fact it can be the drug. This would also increase reporting to the FDA which would make that AER system a bit more accurate.
HorusCat
Laurie,
Your idea makes perfect sense, of course. The problem begins earlier for the company, however. Say drug A has some reports of suicidality. One retrospective analysis supports this. The relative risk is significant, but the absolute risk is still very low. No one is really sure if risk exists at all. It would be great if physicians could warn their patients just to be on the lookout–but then the press gets ahold of it and all of a sudden drug A is killing people by the scores! The company is covering it up! Doctors are being irresponsible!!
The education process is time-consuming and nuanced, but the mass media aren’t concerned with detailed explanations of data collection and relative vs absolute risk. They are concerned with selling copy and advertisements. That is a COI that is never highlighted. No wonder a company tries to sit on a potential problem for as long as possible–publicity is the kiss of death, even if the product is exonerated or the absolute risk is very low.
Furthermore, while you assert that those complaining of a side effect are told that their symptoms are part of the disease process, many times this IS the case. The process of teasing out whether a symptom is due to disease or drug is very difficult and sometimes impossible. SSRIs may initially increase suicidality–and this may be part of the healing process itself (as people begin to emerge from physical and emotional lethargy faster than they start feeling better). Depressed people DO, in fact, commit suicide–and often no one had any inkling they were planning it. The studies looking at suicidality and the SSRIs are confusing and mixed, despite what the SSRI-haters on this site say. All of this is very confusing and complicated even to those who are trained in neuropharmacology, physiology and statistics–imagine trying to explain it to a layperson. Such education must occur, but it would be helpful if it could take place without hysterical media hype.
And after all is said and done, there WILL be bad outcomes with every drug–prescription and over-the-counter. Our society is so accepting of so many risks–driving our cars is much more dangerous than taking an SSRI, as is drinking a few beers and taking Tylenol–and yet it is so intolerant of risk elsewhere. While there is no excuse for a company and/or a doctor not to disclose specific risks with every medication, one would have to be unutterably naive not to realize that every medication has risks and one has to be alert and aware when taking them. No one with an average or above IQ can possibly say that they believed a drug was “perfectly safe.” We all have to accept some responsibility for the choices we make about our health.
Justice in MI
Good discussion here.
Re: corporate branding, Bob will know much more than I. But I recall a study a couple of years ago that suggested patients actually do remember brand and that it made a difference in their thinking about products. I recall the results were a surprise to industry marketing people. But that was only one study.
Re: the impact of corporate structure, I think a number of additional dimensions could be added to Jaynesday’s good list. But all of it leads back to the same essential conclusion (which he makes): That in a context in which the consequences of success, and of error, are so enormous, there is cause for a different kind institutional self-awareness (the impact of structure on decision-making) than is usual, even for the most competitive other industries.
Perhaps pharma will be the industry that discovers the most important and forward-looking model of corporate social responsibility. What a terrific (and, to cynics, ironic) development that would be. Talk about blockbusters!
Laurie
“SSRIs may initially increase suicidality–and this may be part of the healing process itself (as people begin to emerge from physical and emotional lethargy faster than they start feeling better). ”
We’ll have to agree to disagree on this point. But even “if” this is the case, which it may be in some circumstances, warning is still appropriate.
HorusCat
Laurie,
No argument from me. I think patients should be warned about possible mental changes with any drug that crosses the blood-brain barrier, whether the drug is meant to affect mental processes or not. We just don’t know enough about the brain to assume that a drug that passes into it will have no effect on what goes on there.
Jaynesday
JiM,
I also hope that pharma discovers the forward-looking model of corp. social awareness. The impetus for the change will be financial reward.
Even the altruistic environmental “green” movement of today waits for the financial “green” backing of the consumer. The consumer is the force the mandates change.
Bob Freeman
Justice and Jaynesday, one additional comment re CSR and corporate branding. These are very high-level activities at any pharma company. CSR is often a Board of Director level concern and minimally staffed at the global HQ level. Corporate branding is at the country-level but, as I indicated previously, not a priority for divisional staff.
All companies have a mission statement that lists their core principles. I think most employees are aware of these, but that is not CSR. I doubt many employees, including senior ones, are aware of CSR unless they read the annual report or frequent the corporation’s website.
Jaynesday
Bob, re - your comment about CSR and branding - It would seem that there might be some very uncomfortable board meetings in some pharma companies these days then. Isn’t merger of CSR and branding something that happens on a very subtle level? Is it not something that the consumer might not even be aware of on a conscious level? Something like - When does the word Merk instill a sense of distrust rather than trust in the consumers mind?
These things must shake the very foundations of a company. I can imagine that there a number of strategies used to maintain the brand name. Good and bad.
As you can probably tell my understanding of such things is minimal but I find the discussion very interesting.
Justice in MI
There are a few instances of companies (not whole industries) radically transforming themselves. There was a time, as Bob knows well, when a number of pharma companies were, indeed, the most respected of any - Merck most famously.
I agree it will take that kind of revolution. Rationalist economists notwithstanding, there is some pretty good data now that people are willing to pay more for the products of companies whom they know walk the talk.
Of course, as in games of chicken, no one wants to risk being the first. And a number of the factors we’ve mentioned - especially quick turn-over of top management with little or know company memory/commitment - all work against such developments.
CMC guy
Jaynesday you state “Hope I’m not being over dramatic when I say, what a recipe for disaster.” If one only focuses on negative aspects then this may be the conclusion. There are many other elements involved and as has been indicated is complex situation (plus note most of the items are not exclusive to pharma). I am not ignoring problems exists and can be inefficient means but there are so many diverse disciplines required that it is hard to align everything as required. It is much a “checks and balances system” like US government (and often wonder if that currently is a disaster). The following quote comes to mind.
“It has been said that democracy is the worst form of government except all the others that have been tried.”
Winston Churchill
There is mention of “model of corporate social responsibility” and I would argue that this was once largely true in pharma companies so it is more recapturing what has been lost. The pervasive need to maximize profits to keep shareholder happy makes it difficult to return to more altruistic approach.
Chris
Pressure on profits comes not only from shareholders. Management bonuses are tied to a product’s success in the marketplace. This also has an impact on decisionmaking and CSR.
Jaynesday
Good Morning CMC, Justice, Chris I agree with all the points you make here and they are not unique to pharma.
The stakes are high for pharma companies mainly because few products so directly affect the health and well being of its consumers.
Just A Thought
Horus Cat,
I do appreciate your trying to explain the system better. It’s wonderful that none of the doctors you approach have patients with small therapeutic windows. Some doctors are reporting problems. I am not all that “exquisitely unique and sensitive” though that was a kind and thoughtful way to put it.
The FDA and Pfizer have been aware of this problem for months now (as has everyone else we could think to report to). Yes I was on Dilantin. Currently sampling different medications. The “new look” Dilantin is not an option for me. Hoping to hit on another that works while some never find the right AED. I was lucky that Dilantin Kapseals were the right fit for me, right off the bat, and kept me safe and well for years. For now it’s a battle of allergic reactions, partial seizures, and poor control. It makes it impossible to conduct the daily business of life. I’m still hoping something else will work and won’t land me in the ER yet again. My options are limited. Some of these drugs have black box warnings. No one would volunteer for this.
Pfizer doesn’t quite appear to be the thoughtful and caring company that they paint themselves to be. They took away my medication without warning. And based on a one dose study (as I understand it) got it approved without much consideration from the FDA. If it was a mistake, fine. Deal with what you’ve done. But to ignore your post-market study is nothing short of uncaring. We were unwilling victims. Not only that, because we cannot just stop taking our medication, we were captive participants. They kept it quiet.
I have a have a list of people and their dangerous reactions, as unscientific as it may be. The data will be reported again in as thorough a form as I can manage and as my finances allow for copies (isn’t someone already paid to do this?). We received calls and USPS mailings from Pfizer. Form letters in emails from the FDA. Mine came in December of 07, if I remember correctly (I’d have to look).
To switch people to a generic can toss them into a dangerous situation. That stance is the whole basis for the current pending legislation to force insurers to pay for branded AEDs (as per doctor’s orders). Our doctors wrote No Substitutions on our prescriptions and we got the “new look” Dilantin and it caused some people traumatic problems, as could any change in medication. They are not interchangable for everyone. A percentage of people are doing fine on the new formula… but a percentage would do fine on any generic as well.
We are the ones that this legislation was designed to protect… the “exquisitely unique and sensitive”.
Did you read Ed’s article?
http://www.pharmalot.com/2008/05/a-new-version-of-dilantin-is-giving-pfizer-fits/
Just A Thought
have a have a
:-/