Explain this bizarre CVS Caremark prescription policy

I have Caremark prescription insurance through my wife’s employer. My doctor prescribes my medication requiring brand name. He says the dosage in generics isn’t well controlled. For some reason, today Caremark called me to get me to authorize the brand name, because they would be charging me a “DAW penalty” (“dispense as written”) of $63. Huh? After a couple more conversations they told me that if the doctor specifies “dispense as written” they charge the full price for the medication. However, if he allows a generic substitute, they fill it with the brand name anyway but they only charge me as if they were providing the generic.

It makes no sense to me that they charge different prices for the same plan, same patient, same doctor, same medication–the only difference is the magic words he writes on the prescription pad. I don’t know if this is Caremark policy, or a consequence of the plan specific to my wife’s company.

I got the same story from two different people so I got the sense this was correct, but nobody could explain it in a way that made it sound like anything but nonsense.

Any industry insiders understand this?

All of the prescription drug plans I’ve had had different levels of coverage for different drugs. There’s one copay for a prescription that allows generics, another for prescriptions written “dispense as written” , there are drugs that aren’t covered until I’ve tried some other, less expensive drug first and so on. Some drugs require me to pay not only the brand-name copay but also the difference in price between the generic and the brand name.

But here’s the thing- it’s always been characterized as “when there is no generic available” or “when the prescription does not allow substitution” or “when you choose the brand name”. It’s never been “if the pharmacy doesn’t stock a generic version”. It’s entirely possible that there’s some contractual provision that requires Caremark to eat the difference if it’s their decision to fill the prescription with the brand name.

I do think that’s correct- if the doctor allows a generic but the pharmacy chooses not to fill it, why should the member be penalized?

(As a side note, in my experience CVS Caremark is one of the hardest PBMs do deal with on a professional level. Prime Therapeutics and Express Scripts are a much easier both with direct contacts and to use their IT systems.)

There are different types of DAW codes, the most common ones are:
[li]DAW-0: No Dispense as written (substitution allowed)[/li][li]DAW-1: Brand Required by Prescriber[/li][li]DAW-2: Brand requested by patient[/li][li]DAW-3: Brand selected by pharmacist[/li][li]DAW-4: Generic not in stock[/li][li]DAW-5: Brand dispensed as generic[/li][li]DAW-7: Substitution not allowed - Brand mandated by law[/li][li]DAW-8: Generic not available in marketplace[/li][/ul]

The requirements to use a DAW-1 are pretty strict, and are rooted in law for the state in question. In my state (Georgia), the requirement is the magic words “Brand Necessary” or “Brand Medically Necessary” written on the Rx. Anything else the doctor puts like “no substitution”, or signing on the line that says dispense as written, or putting the letters “DAW” in a box, or whatever else a doctor might put would be submitted as a DAW-2.

I have seen insurance companies have different copays depending on the DAW submitted. Most of the time they will give you the normal brand copay with a DAW of 1 or 7, and a penalty with a DAW-2, however, sometimes they won’t cover a DAW-1 without a prior authorization (when the prescriber calls and convinces the insurance that you REALLY need brand, which might be hard if the insurance doesn’t agree with the prescriber). One other thing, if they offered to just send you the generic, then it might not meet legal requirements and they are submitting it as a DAW-2, if it is a true DAW-1 they would have to get the Rx changed from the doctor.

Now, one other thing it could be… There are what is called authorized generics, where the brand company will produce a generic drug of their own brand. It is normally the exact same pills, from the same line, but sold as a generic. It is possible that the generic that Caremark is using happens to be the authorized generic for that brand, and the person you spoke to knows that, and was just trying to get you to accept the generic knowing it was the exact same pills. Though, it could then change in the future if they change their preferred generic.

Mind if I ask what drug it is? There are only a few drugs that brand vs. generic actually matter. From what you posted it sounds like you are dealing with mail order, so it is most likely a 90 day supply. If you were really paying the full cost, the only brand that would cost around $63 for a 90 day supply that I can think of would be Synthroid.

Imagine this

Two people use the same Caremark policy, and live in the same street, having the same illness, the same doctor , the same supplier, the same office, on the same day, and we assume for the time being the generic is totally as high quality and effective as the original brand. … All other things being equal…
person 1. Doctor ticks “no substitution allowed” and customer authorizes this with Caremark, and pays the $69 DAW penalty … and get the original brand.

person 2. Customer arranges that substitutions are allowed, pays no DAW penalty and get the original brand.

Does that ever happen ? They get the same thing but person 1 paid $69 DAW ?

If it does happen, who lost out on the supply ?
Does Caremark pay the higher price to the supplier ?
Or is it the supplier dropping the price ? price matching ? (thus admitting they keep a fake list price for the purpose of gouging people like person 1 !)

Opposite of my situation, review OP.

Based only on your list, they are charging me extra for DAW-1, but are telling me that if the prescription is written so that they can code it as DAW-5 (“brand dispensed as generic” are the words they used when they tried to explain this to me) they will charge me the co-pay of something like $20.


Can a doctor write “DAW-5” on a prescription, or is that purely an internal coding issued or controlled by the insurance company?

Speaking of synthroid, why are its generic versions so unpredictable? I had similar problems a few years ago until I read the formulary and found my insurance would cover name-brand Levoxyl at the generic co-pay.

odd, my wife encountered some problems getting her synthroid scrip filled at CVS, too. Took a trip back to the MD to get something changed on the scrip. I’ll ask her to take a look at this and comment.

On my Caremark plan, I pay $12.50 for generics and $62.50 for nongenerics. It sounds like you were just paying the usual price for a nongeneric and the person calling just explained it wrong (not unusual for Caremark – we had so many screwups from them that the VP for customer service had to get involved and we were assigned a special rep to check on things).

I think that was in response to this line in my post

which I suspect is at least part of the reason for the different DAW codes. If all insurance companies cared only about whether the customer got the brand name or generic and not the reason why , those DAW codes wouldn’t be needed.

I have been paying something like $20 for Levoxyl for at least 15 years. Then they quit making the dosage I need (137 μg) and the doc switched to Synthroid, and that’s when this problem started.

It is not so clear that they are. The FDA’s take:

Nevertheless there are persistent concerns that generic and name-brand may not be exactly the same … although for most people the actual practical clinical importance of any difference, if any exists, is also unclear.

Is Synthroid the only possible substitute for Levoxyl ? OP quoted Caremark as suggesting
"You’d get Synthroid anyway, so why bother paying a DAW penalty for the tick on the prescription saying “No substitutes.” ?? OP would accept Levoxyl and its clearly the equivalent given the explanation of the change to Synthroid due to the difficulty with SUPPLY of Levoxyl .

Well anyway the issue with supply of Levoxyl is surely no reason to say “no substitutes”.
The substitute would be a fine source of 137 cmg Levoxyl equivalent.

If Levoxyl was good enough, a generic of it is also good enough. Only synthroid had the higher quality.

My other question remains… but I guess the OP already answered it in the OP… The insurance company charges the DAW everytime the doctor ticks the box.

I figured it was Synthroid, it is the one that gets all the talk about brand vs. generic, and is the only real brand that someone would actually pay full cost for… One thing I will say, I don’t see any reason they would actually dispense brand Synthroid for the cost of the generic (your talking generic at my chain, $15.99 for 90 days, vs. brand $60+)

No, a doctor can’t write DAW-5. The DAW codes are a coding issue between the pharmacy and the insurance company (and our software of course). I could put any DAW code on any prescription. But once the insurance company does an audit, if I didn’t do it properly they will do a charge back, for the ENTIRE amount of the claim. For the amount they paid us PLUS the copay… I.e. We just gave the drug to the patient for absolutely nothing.

The insurance company doesn’t care with what the patient got, it only cares with what they need to pay for, and what excuse they can make not to pay for it.

See below.

See below.

Ok, Synthroid is one of the few drugs that brand vs. generic actually matters. The reason it matters is because levothyroxine has what is called a narrow therapeutic index. There is a slight difference allowed in generics vs. brand; in most cases it doesn’t matter. You’re prescribed 20mg of Lipitor, the generic gives you 19mg or 21mg, it makes no clinical difference, and no one could tell. Drugs with narrow therapeutic indexes, like Synthroid, you’re prescribed 100mcg, you get 95mcg or 105 mcg, it might actually matter… BUT… drugs that have narrow therapeutic indexes are monitored really closely, and we will adjust the dose based off you, the patient, not the drug…

So, what really matters, is that what you start with, is what you continue with… With Synthroid, if you got stabilized with the brand, you stay with the brand, if you stabilized with the generic you stay with the generic… And don’t change generics! Stay with the same manufacturer! Levothyroxine is one of the few drugs my chain doesn’t change manufactures willy-nilly, mainly because they know it matters. The main problem is when people change pharmacies, that use difference manufactures. That slight difference actually matters in this drug. Changing manufactures will put you back in the whole titration period… This was a big issue a few months back, when brand Levoxyl was recalled, was unavailable, and everyone had to switch people to another generic.

For the record, the only drugs where brand vs. generic REALLY matters is thyroid drugs, blood thinners, anti-seizure drugs, and some antidepressants (Mainly Wellbutrin XL).

Not accepted…

Generics are fine… here’s a report which is saying that the generic is fine, just as good.
And Boot’s guilty behaviour when the report didn’t give the results they liked ?


Boots thesesdays try to recommend against changing brand, because “the effectiveness may vary”. …
but why can’t OP change to another cheap brand ?

Hika T’Bawa,

I am not quite getting either why many (apparently including the op’s physician but it is claimed all over the place, including by the endocrine societies) believe that thyroid hormone replacement has such an exceptionally narrow therapeutic index.

Also the FDA at least contradicts your claim that the between product bioequivalence is significant, stating that the within manufacturer product variability based on degradation over time was (and past tense claimed) the issue.

As to the former - the vast majority of those with hypothyroidism have some significant remaining thyroid reserve. As pointed out in the citation I provided that found a difference, that difference was potentially significant only for those without such reserve, such as those with congenital hypothyroidism, not for the bread and butter acquired hypothyroism (e.g.Hashimoto’s). Even for those with congenital hypothyroidism the difference was of borderline clinical importance: thyroid hormone levels were not impacted; only TSH levels. But to the point, having thyroid reserve should mean that the therapeutic index is not actually so narrow.

In short, what is the basis for believing that 5 to 10% difference in potency of levothyroxine matters so much more than 5 to 10% potency difference in a statin or an antihypertensive or an OCP …? What is the evidence that even borderline high TSH with normal thyroid hormone levels is better or worse for you than borderline low TSH with normal thyroid hormone levels?

Your bottom-line point however seems most reasonable. IF one believes that there is a risk of a consistent (and significant) inter-manufacturer difference (which is what is suggested by some), then picking the generic version, prove that that your TSH is stable on it 6 weeks later, and save money over the decades, makes good sense. Staying with the much more expensive brand name makes no sense at all.

BTW, any sense of if and when albuterol hfas will be available generic?

To be honest, even with the study DSeid linked, I’m still not convinced that this is necessarily/universally true. Currently, we treat a change from one generic to another or from brand to generic or vice-versa as if it matters, yes, but the data on the subject is less convincing the deeper one dives into it (IMO). This study looks like it might establish that, at least in certain subsets of those we give thyroid hormone replacement to, there is something more concrete to the current recommendation regarding switching. A single study, however, in what arguably isn’t a very generalizable population, with a very low number of involved patients (31 patients total, 16 originally receiving Synthroid, 6 on Levoxyl, and 9 on unspecified L-T4 generics, all stabilized at entry), which appears to have only provided comparison statistics on Synthroid vs the Sandoz generic or Synthroid to Synthroid (corrected for if the 8 week trial of the generic came first or last), but no mention of Levoxyl to generic or Synthroid, or unspecified generic to Sandoz, probably needs replicated before I’m comfortable stating that brand to generic provably does matter. It wouldn’t hurt if the replication was powered to detect differences going from Levoxyl (if/when it is again available) to Synthroid or generic, or one generic to another, if they really want this to validate a difference. I’ll also note, judging by Figure 1 in this study, that even with an apparently statistically significant change from brand to generic in Congenital Hypothyroidism, the variance in most of the patient cases looks like it still left the patient in the reference adult normal TSH range of 0.34 to 5.6 mU/L, which leaves me wondering, even if the statistically significant difference holds in further studies, what the actual clinical significance will be.

Nitpick: the allowable differences apply between batch/lot for both brand and generic products, as do drug potency requirements across the stated shelf-life of a given lot of the drug. In the case of levothyroxine sodium products, the allowable potency degradation over the shelf-life of the drug is a decrease of 5% (range 95-105%, up to 105% as an artifact of manufacturing and assay variabilities, not as intentional formulation overages to meet the requirements). For those keeping track at home, in the linked study in DSeid’s post, 25 mcg tablets were used, so IF they degraded the full 5%, we’re talking a decrease of 1.25mcg per tablet max in order to meet current FDA standards. Given that the children in the study were recruited from 2006 to March 2010, and the final rule narrowing the potency difference from 10% to 5% didn’t go into full effect until 2009, we may actually still be seeing an artifact of when intra-brand potency issues still were prevalent. See this FDA report for more info on the potency issues which prompted the rule change. And even if the tablets degraded more, that may be an artifact of poor storage at home, given how rigid the potency testing/storage is in the manufacturing setting (25 degrees Celsius +/- 5, Relative Humidity of 60% +/- 5%, each batch tested coming from sealed, light blocking bottles with a dessicant) versus what you’ll find at home where people STILL store their medications in a bathroom (heat and humidity are the enemy, in the case of most drugs).

Of course, potency is just one part of the brand vs generic debate and is usually the one, in my experience, most often mistakenly conflated with the rules governing what is and what isn’t bioequivalent. In other words, when a patient, prescriber, or sometimes sadly, even a pharmacist, states that a generic could have as low as 80% of the amount of drug in a branded product or as high as 125% (a 45% range), and thus use that as an excuse to say that brand is better, they are, in fact, wrong (note, I am NOT saying that this is what Hirka was saying, rather this is what I most commonly run into in my own professional experience). Every drug, save levothyroxine, must fall within 90-110% of their stated potency with no greater than 10% loss over the drug shelf-life, while what is required of levothyroxine as noted above.

The 80-125% thing is actually a misunderstanding of statistics. In order to meet bioequivalent standards, in addition to falling within the rated potency/dissolution standards, a generic must show that it’s Cmax (the maximum concentration it reaches in the blood) and it’s AUC (Area Under the Curve, a measure of total exposure to a drug from time zero to extrapolated time infinity) meet a certain standard, in ratio to what happens with a reference drug in the same person (usually a healthy volunteer). Those ratios must fall within a 90% Confidence Interval of 0.8 to 1.25 (80-125% of the reference drug, or 4/5 to it’s inverse 5/4 for those wondering why only 20% the one direction, but 25% the other). I’ll let dopers who are far better versed in statistics try to explain the difference, because I (think I) know what it means, but I can’t explain the concept of a confidence interval to save my life. :stuck_out_tongue:

Now, fun fact time: there are currently four reference drugs for levothyroxine sodium products. Unithroid has been designated as AB1, Synthroid AB2, Levoxyl AB3, and Levothroid AB4. Unithroid is manufactured by Jerome Stevens Pharmaceuticals for Lannett, who not coincidentally also sells a generic of this product which is also manufactured by Jerome Stevens Pharmaceuticals and comes with the same imprint, same formulation, and same coloring as the brand. This product has been deemed bioequivalent of Synthroid and Levoxyl, but not Levothroid. In point of fact, the only generic which can state that it is bioequivalent to all 4 reference drugs is Mylan’s generic. In this retrospective study, published in the same journal and issue as DSeid’s linked study, the authors looked at patients who had been treated with only brand or generic levothyroxine products in the age range of 0-36 months (the range where L-T4 is going to most impact neural development) and found no difference with one statistical test between groups, and slightly better control of congenital hypothyroidism with generic using a different statistical model, showing, at least retrospectively, that those thought to be the most sensitive to L-T4 (aged 0-36 months with congenital hypothyroidism) can be treated with $4 generic levothyroxine (Lannett and Mylan were the two named generics, though the majority of the generics were either multi-sourced or unidentified per the authors) and have as good and possibly better control outcomes vs brand Synthroid.

Sorry, I disagree, particularly in the case of Wellbutrin XL (it was one strength of one formulation of one generic manufacturer and did not apply to any of the other generics on the market, iirc). Some argue that it matters, but these are usually the same patients and prescribers mis-quoting bioequivalence and potency standards, and often, anecdotally, in my experience, are also on drugs which may affect gastrointestinal motility and/or stomach pH levels which may throw things off slightly from brand to generic (and should be termed more correctly as a drug-drug interaction), or important dietary factors (excess vitamin K1 ingestion for Warfarin) which are overlooked/missed.

You’ll probably enjoy this review article from Bolton, published in 2005. It largely debunks the current Endocrine Societies’ arguments about brand versus generic, including the claim made by the authors of your previous cite about the difference of 12.5% being undetectable even with background levels removed from Blakesly et al, which was, shock and amazement, funded by Abbott Labs (now Abbvie), the makers of Synthroid. In essence, Blakesly et al wasn’t powered properly to detect the difference because the variability in background levels left a greater impact of noise on the 400mcg and 450mcg doses than it did on the FDA’s current standard of 600mcg (and this was confirmed in another study, quoted in the review article, using 500mcg and finding the correct ratio of 600mcg/500mcg in both Cmax and AUC of 1.2).

Bolton above pretty much agrees, stating:

According to the Code of Federal Regulations 21 CFR 320.33, narrow therapeutic ratio (NTR) is defined as follows: “There is less than 2 fold difference in median lethal dose (LD50) and median effective dose (ED50), or there is less than 2 fold difference in minimum toxic concentrations and (LD50) and median effective concentrations in the blood.”11 However, there is no scientific evidence for this assertion with regard to levothyroxine.

He further goes on to note that missing a single dose in 1 week would lower blood levels by more than 10%, and patients aren’t always 100% compliant, so we should have a heck of a lot more uncontrolled hypothyroid patients if the 12.5% Blakesly et al claim were true.

Sadly, based on my reading, mostly theorized FUD, targetted largely at three populations: 0-36 month old children due to the risks of low L-T4 on the developing brain, patients who might be at risk for heart issues if they go into a mildly hyperthyroid state, and patients who are being treated aggresively for Thyroid Cancers.

I haven’t heard anything, and Pharmacists’ Letter’s upcoming generic lists doesn’t list anything, so sadly, I’m guessing it might still be awhile. :frowning:

Thank you for the link. Yes, interesting.

I actually am receptive to the argument that TSH levels are important and that significant differential impact on TSH levels would argue against bioequivalence. How much difference would constitute practical clinical significance I am unsure. And I can see the concern for a few specific high risk populations as you’ve listed who either or both have little thyroid reserve and indeed are at higher risk for adverse effects of being outside of target range. The insistence of many docs on DAW for the typical acquired hypothyroism patient however seems to be more a score for detailing than for critical evaluation.

Agreed on all counts. I won’t rule out TSH potentially being an important parameter given that outside of central hypothyroidism (where Free T4 is a more important measure, if I’m remembering the AACE guidelines correctly) TSH is the more important parameter clinically. Heck, the model used in this study by Eisenberg and Distefano (the authors state that it has been clinically validated, and I’m certainly not equipped to say otherwise) may actually provide support for the difference in response from Synthroid to generic in the earlier linked Carswell et al study, since they take the 150mcg tablet data from the earlier linked FDA report on intra-brand deviation over the shelf-life of L-T4 products and use simulated thyroidectomized patients with no endogenous T4 pool and determine that TSH falls within the selected range of 0.5-2.5 mU/L with the 96-104% potency range, barely fails at 6 months for that range with 95-105%, but fails in the prior 90-110% potency range. They also determine that a drop in absorption resulting in a change from a bioavailability of 88% down to 79% can affect the same change. Given that only 7 different lots were used (and expiration dates not listed, nor how close the product was to expiration) of Synthroid in Carswell vs 12 lots of generic Sandoz, it may just be that the authors happened to get Synthroid with relatively low potency degradation while (again based on the time course of patient enrollment) the 12 lots of Sandoz might have included more close to expiration 90-110% range pills. Also, since about 2009, anecdotally, the Synthroid products we receive in the pharmacy I currently practice in all tend to have less than 1 year expirations on them when we get them, which may be Abbvie narrowing the expiration window intentionally to meet the FDA potency requirements, rather than reformulating their product.

That being said, the endocrine societies’ call for TSH measurement as part of bioequivalence standards could set off a flood of problems for the FDA if adopted, since no other drug has to show any form of downstream effect in order to demonstrate bioequivalence currently. If you add TSH for levothyroxine, you’ll get calls from companies like Pfizer urging the FDA to consider LDL-C reduction as the only sure-fire way of proving Apotex generic atorvastatin is truely bioequivalent to Brand Lipitor (or Approved generic Greenstone), which is a road I don’t think we need to go down.