How so? A new study shows that patients who get samples end up with significantly higher out-of-pocket costs than those who don’t. On average, patients who left their doctor's office with a tidy stash wound up spending 37 percent more for their meds during the six months they received samples, and 19 percent more in the six months afterward, than those who didn’t get samples, according to the study in Medical Care.
To be specific, patients who never received free samples spent an estimated $178 out-of-pocket on drugs over six months. By comparison, those given samples spent $166 during the six months before receiving samples. But they spent $244 during the six months they did receive samples - and $212 in the six months that followed, according to the study, which examined 5,709 patients from the Medical Expenditure Panel Survey, which is government data.
"There are two conflicting theories why out-of-pocket costs remained high," Amirban Basu, a University of Chicago professor of medicine and co-author, tells us. "One, people may be sicker and continued to need medication. So they're getting samples, but continue to spend. The second possibility is that brand-name promotion induces them to continue spending. This would undercut the conventional belief that physicians help all patients by giving them samples."
The study also found that younger patients who are not on Medicaid are more likely to receive free samples, and that patients on Medicaid were less likely to receive samples than patients with other types of insurance. This may, in part, reflect the targeted detailed effort carried out by the industry, Basu adds.
14 Comments
Not exactly earth-shattering, and limited by it's retrospective design. Ideally, I'd like to see a prospective, randomized study that compared total medical costs between two groups with an identical diagnosis and identical script (or at least a script for two drugs in the same class), but one group got free samples and the other didn't, and then compared the total out-of-pocket medical costs between the two groups. That would nicely isolate the variable (here, samples).
In this study: A) Patients who got samples spent more money than those who didn't. Of course patients who got samples also probably just started a new prescription - whereas patients in the control group did not necessarily just start a new script.
B) Second, patients in the sample group were significantly less likely to have insurance than the control group. People in the group less likely to have insurance (the sample group) faced higher out-of-pocket costs than people with insurance. Not a shocker.
The authors clearly covers the limitations of their study from point A, and at least mention point B (although I didn't see this in the limitations and it seems like a VERY big deal to me). But of course the media (here Pharmalot) fails to really capture these important methodological differences with the catchy headline:
Dig Deep: Those Free Samples May Cost You
I'm sorry but this kind of study really ticks me off. What is the value of reporting such banality without identifying any form of causal relationships. Is there any evidence here that having samples causes the subsequent increase in OOPs? No. And as to the so-called "two conflicting theories", I can think of so many more than two theories as to why this should be occurring. Trying to link these data to DTC or detailing is ludicrous. Did the author think that just perhaps doctors gave more samples to younger patients because Medicaid patients don't have to pay so much for their drugs - because they are eligible for Medicaid!
I'm sorry, but this looks like another example of academics trying to find a way getting press coverage by attempting to show that the industry is ripping people off by using distortional promotion tactics related to sampling, when in fact the data show no such thing.
Hi Jack2 and Outside,
Yes, the study has its limitations, but these were largely noted. And of course, headlines have limitations, too. Space is the issue. However, if you look closely, I deliberately used a qualifer - the word 'may.' I didn't write that samples will definitely cost anyone more.
Meanwhile, I've written the lead author and asked him to respond to your comments. Let's see if or what he does.
Regards, ed
I don't think there's anything wrong with the study per se. Look at the title: Characteristics of Patients Receiving Pharmaceutical Samples and Association Between Sample Receipt and Out-of-Pocket Prescription Costs
This study has limitations, but all studies have limitations and it would cost far, far more to run the prospective study I described, and by the time you finished it all the drugs would be available generic, making your results fairly meaningless.
I also concede that writing a brief headline is difficult. I just think it's equally possible that free samples, given to un/under-insured patients could: A) actually decrease out-of-pocket costs for these patients. B) address the concern that some people have with drug-costs (and really all healthcare costs since drugs only represent about 10% of total costs)- which is that they need to pay for the drug whether it works or not. This is really a try before you buy situation for people without insurance. No one can make them go refill the script.
Ed
No problem with you at all, I just think studies like this - especially with the "interpretive" comments provided by the authors - give ample fodder to the industry bashers without actually adding anything meaningful to the body of knowledge.
Jack2
I agree, but it would have been useful if the title had "Some" at the start.
Hi Jack2 and Outside,
You both made good points. And I appreciate the insights. I don't pretend to be astute as others in the fine art of analyzing studies, but do like to offer them so dissection and discussion can add as much clarity as possible. In that way, perhaps we can all learn still more. But that's just a hope.
Anyway, thanks for the input. ed
Although identifying casual relationship is generally hard (even with a randomized trial in this context due to the logistics of implementing it), identifying associations and sensitivity of those associations to various observed confounder is generally very useful. (Otherwise the whole field of epidemiology research would be thought to be a banality).
Spending on samples far exceeds that on detailing and DTCA and so to think that pharma's motif to supply free samples is purely altruistic is ludicrous.
We are cautious in not drawing that link and believe have fairly represented the alternate mechanisms of action in the paper.
We agree that patients on Medicaid have more generous coverage and so may be less likely to receive free samples. However, we do find: 1. Uninsured were not significantly more likely to receive samples than the privately insured. 2. After controlling for insurance and income, older patient were significantly less likely to get samples than younger patients. 3. All our results on out-of-pocket costs are adjusted for differences in insurance status. 4. We see no significant differences in the mean number of prescriptions between the samples group and the control group.
Given these, it was interesting to find that out-of-pocket cost and also total RX expenditures remain significantly high even after they stop receiving samples. We do not rule out the possibility that unobserved severity of illness may drive this result, but nevertheless our results (which is the first time these are documented empirically despite having long lasting debates on this issue) make it more important to tease out these differences for effective public policy.
A little known fact: Branded samples are the number one influencer of the prescribing habits of the health care provider in posession of these samples. Number two influencer, by the way and of no suprise, is gifting, because we can't say bribery, because that's a federal offense.
My feelings are mixed on the sample issue. They have the ability to provide immediate benefit, but they are expensive. Read recently about a company selling generic dispensers to be placed in some doctor's offices, which would be ideal vs. branded samples, if generics exist for a particular brand, as generics are bioequivalent and are being utilized more as time progresses.
Hatch/Waxman Act of 1984. A break for generics. No longer required to conduct multiple studies as in the past, with the exception of one showing bioequivalence to the soon to be unpatented branded med. Branded pharma responded by filing frivilous lawsuits about patent infringement,and such cases can last 3 years. In addition, they bribe generic companies to delay the release of thier generics tat are now a threat to a particular pharma company.
And just when you thought those rich guys were ethical and honest......
From the discussion of the paper: In this nationally representative longitudinal study, we found differences in the likelihood of pharmaceutical sample receipt based on patient’s age, income, and INSURANCE STATUS. [emphasis mine]
How was the calculation that pharma spends more on samples than advertising done? Is it based on the assumption that every single sampled script, if not sampled, would have resulted in the sale of a full price script (the way to get the highest possible cost), or the cost to manufacture the samples (the lowest possible cost), or some compromise in between? The former is based on the false assumption that everyone who got the drug for free would have bought it (It's the same false logic used by the music industry when it claims that every downloaded song is a lost sale).
From the patient perspective I think the samples are great for patients who can get them. I think the primary economic driver for companies to give docs samples don't delve down to the individual patient level. The primary benefits for the company are: 1) it gives the doc a tangible reason to see the rep, and thus the doc will listen to the rep's schpeel. 2) it influences doc prescribing pattern. If it was done to get the patient "hooked" as alluded to in the post, then why would companies provide samples of antibiotics? You could get Z-pack samples. That's a drug you take for a while and your done with it - you're not gonna refill it. Reps gave docs Z-pack samples to get the DOCS comfortable with the Z-pack, so they would prescribe it.
Another reason why drug costs could be higher for sampled patients is that those patients received a brand name drug (since, to my knowledge, only brand name drugs are sampled). Control patients could have received a generic.
Reply to Jack2's comments:
INSURANCE STATUS. [emphasis mine] - We control for these differences in a regression, and so the OOP results are not driven by difference in insurance status.
How was the calculation that pharma spends more on samples than advertising done? - We did not do these calculations. Look at :
Prescription drug trends. Kaiser Family Foundation. November 2005. Fact Sheet 3057–04.
Medical Marketing & Media, “The IMS Health Report – Pressure Zone,” May 2005, Fig. 7 and Fig. 9, at http://offlinehbpl.hbpl.co.uk/Misc/MMM/Features/MAY05%2036-50%20IMS.pdf; Kaiser Family Foundation, Prescription Drug Trends, a chartbook, July 2000, Ex. 3.13, at http://www.kff.org/rxdrugs/3019-index.cfm.
From the patient perspective I think the samples are great for patients who can get them.
- We agree.
I think the primary economic driver for companies to give docs samples don’t delve down to the individual patient level.
- Marketing is all about individual customers and their behavior. It is hard to belive the giving free drug samples is not a marketing tool.
The primary benefits for the company are: 1) it gives the doc a tangible reason to see the rep, and thus the doc will listen to the rep’s schpeel.
- Again, we agree
2) it influences doc prescribing pattern.
- prescribing pattern is also driven by patient preferences which may be influenced by adversting. If it was done to get the patient “hooked” as alluded to in the post, then why would companies provide samples of antibiotics? You could get Z-pack samples. That’s a drug you take for a while and your done with it - you’re not gonna refill it. Reps gave docs Z-pack samples to get the DOCS comfortable with the Z-pack, so they would prescribe it.
- almost all patients are repeat customers of antibiotics. Infections do occur more than once in the same patient.
Another reason why drug costs could be higher for sampled patients is that those patients received a brand name drug (since, to my knowledge, only brand name drugs are sampled). Control patients could have received a generic.
- could be, but does not explain why we see differences in costs in future periods when no sample is received, unless there is inducement.
Again, we stress that unobserved serevity of illness may be driving the results and that is why it is important to continue to look at this issue.
I think this is a very complicated issue, which the authors have done a good job of beginning to look at. Certainly, samples induce doctors to use brand name rather than generic. I know that doctors consider whether a pt has insurance and if so, whether the insurance will cover the brand, before they prescribe. What doesn't get said here is that the doctor may believe that branded is better--it isn't all just pharma seduction. The prevailing mentality seems to be that generic is always better, and I don't think that's the case.
I have had docs tell me that they won't write my drug without samples. Their reasoning is that since these are psych drugs, the patient may or may not respond to them (it's not like writing a statin, where one is pretty much like the next). If the patient doesn't respond or has a nuisance side effect and wants a different drug, they don't want the patient to have wasted a co-pay. Related to this is the fact that I think it is possible at times that docs reach for a more expensive branded drug over a less expensive branded drug because of what samples they have available.
And Dan, you keep throwing this bribery word out. Let's see some evidence to back it up; otherwise, it's just meaningless bashing. What sort of payoff am I supposed to be giving my docs to get them to prescribe? A pen? An $8 lunch?
And the lawsuits you talk about are usually filed when a generic company tries to break the patent BEFORE the drug is due to go off patent. Lipitor doesn't have LOE for a couple of years, but generic companies are already trying to break the patent. What is Pfizer supposed to do, sit back and have their intellectual property rights trampled? As for bribing the generic companies...exactly how does that work again? Your proof would be.....???
Whatever you give a health care provider that has been proven to increase business is not a gift, because reciprocity is expected and likely. The exchange, however mild, is bribery, because of the anticipated exchange that should occur. Gifting, on the other hand, is when a doctor eliminates any elements of obligation and reciprocity to return from that doctor for that gift. Take it a step further. Let's say that you just paid for a lunch at a doctor's office, which is perfectly legal. But for hypothetical reasons, let's say that after that lunch, your numbers are down dramatically, so you tell that doctor that you expect more scripts for your promoted products because of the lunch you bought, so the doctor owes you. Your statement then has manifested into a federal offence. It's called quid pro quo and is in violation of the federal anti-kickback statute. If you refuse to provide services to a doctor of any nature for that matter that have received things from you in the past, that possibly could be viewed as extortion, after such a conversation as the lunch scenario, which has actually happened by some idiot rep.
And just so you know, the study aids you use for your products are largely fabricated and embellished with the efficacy claims. You have this material because the producers of the material does not always go through DDMAC, and only take action when they get busted with a warning letter. If it is a trial on one of your meds that you may be utilizing with your prescribers, someone from your company paid a journal to publish a trial on your med and this trial, statistially is completely inaccurate and without merit, as itobjectivity and is embellished as well. It was published in a journal because a big pharma company paid that journal for advertisements and thousands of reprints if they would publish this flawed study. Normally, the journal will agree to this.
Please correct me if I'm wrong, but perhaps I have given you the red pill. Try and keep an open mind and research claims made from what may seem like such information is presented in these comments from a rather bitter ex big pharma rep, yet I can assure you I am not bitter. I wish to expose the truth about the pharmaceutical industry for th sole benefit to possibly educate others.
Dan, I understand where you are coming from. There are several things to consider: 1. The "bribery" you speak of occurs in every aspect of business in a free-market economy, from our politicians with lobbyists, to your local government official taking a pen from a contractor, and so on. People seem to think that somehow, medicine should be insulated from this process--and that isn't going to happen in a society that is based on free enterprise, entrepreneurship and a market system. You can argue for complete government funding of medical research. I would strenuously disagree with that proposition, but that is a discussion for elsewhere.
2. You know yourself that doctors are incredibly hard-headed, and the reps that try that little trick of "now you owe me some scrips" are likely to get kicked out of offices. You are right--it is technically illegal, but more than that, it is an incredibly stupid marketing technique practiced by few reps.
3. You go overboard on the study thing. Who do you think will do the clinical research if it isn't the people who have skin in the game--the drug companies. Again, this is a free-market economy, not a fascist state where the government takes care of everything. Your argument breaks down because it is so generalizing. Do the companies influence how the trials are written up? Yes. Do they falsify data? Not on any large scale. Just think about it. If a company claimed that drug x works on bacteria y, and produced a trial to that effect, and docs then prescribed drug x to treat patients with bacteria y--if the drug didn't work and the patients stayed sick, then docs would quit prescribing the drug. Likewise, to take a real life example, if the drug I sell for an auto-immune disease didn't produce relief from the symptoms, then docs would quit using it. And as I talk with my docs, we freely discuss the limitations of the medicines we have and the terrible need for the development of more effective medications.
3. I never use my visual aids.
4. I can only speak for myself, but I have conversations with doctors about real world issues with my drugs, the disease states they treat, the limitations of the meds, the very real benefits they present, and basic problem-solving things like dosing a drug differently to avoid a certain side effect. It's not rocket science, but I think it is helpful to them.
Like I said, we live in a free society which depends on capitalism and a regulated free-market system. It ain't great--but it's better than anything else that has ever been tried. We need to free ourselves of the mindset that somehow "medicine should be different" because that is the path to stagnation. You want to throw the baby out with the bathwater.
Dan, You know what is ironic? We are prohibited from providing really useful information, like which pharmacy has the cheapest drugs. It is technically illegal for me to tell a doc that a local grocery store is now offering $4 generics, just like Wal-Mart. How stupid is that?