Defining the Impact of AIDS Drug Pricing on the Public Sector
Examining the Pharmacoeconomics of U.S. AIDS Drug Access, Washington, DC, April 6, 2004
I wish to thank the International Association of Physicians in AIDS Care (IAPAC) for inviting me to join you all here today to discuss the impact of AIDS drug pricing on the public sector.
My estimate of expenditures for antiretroviral drugs in the year 2003 in New York State, by payer, is shown here (Slide 1). There is some guesswork involved in this, however. The one thing I am fairly certain of is that the ratio of Medicaid to AIDS Drug Assistance Program (ADAP) is just a little over three people on Medicaid getting antiretroviral drugs for every one person on ADAP. Based on recent corrections to some previous estimates that we had, and more recent data that we picked up from the Veterans Administration (VA), we are pretty confident that these numbers are pretty close, although there are still some estimates figured in. The biggest guess is with the "Private/Other" category. We lumped a number of smaller things in there so its accuracy is least certain, although I think it is still quite close.
So, New York State spent somewhere over US$800 million in 2003 on just antiretroviral drugs. This cost, spread across various programs such as Medicaid, ADAP, and the VA, is not the same mix that you are going to find in every state. Medicaid in New York is a very inclusive program that covers a high number of people in need of antiretroviral drugs. Christine [Lubinski] has said, "Medicaid is not universal healthcare for the poor." In New York, it is. You do not have to be disabled to get it; you just have to be poor. In some states, the ratio of Medicaid to ADAP will actually reverse, because Medicaid is such a closed system and so difficult to get into that ADAP will have to pick up that much more of the burden.
Here is where we get into the detail. Just for perspective, I think that the antiretroviral drug market in the United States is about US$4 billion per year, and ADAP and Medicaid are probably good for at least US$250 million and US$1 billion of that, respectively. There are a mix of government prices and purchasing systems. The purchasing system used in a distribution system to some extent dictates how you are going to get your pricing. There is the Federal Ceiling Price, there is a Federal Supply Schedule, and there is a Public Health Service 340B price, for which ADAPs are eligible. All of these mean you take possession of the drugs, you get the drugs, and then you distribute them.
The other primary type of system is the pharmacy network system, where you reimburse a broad network of pharmacies at average wholesale price (AWP), typically minus some percentage, plus a dispensing fee. That is the system that Medicaid uses. Medicaid also receives rebates under Omnibus Budget Reconciliation Act (OBRA) legislation, and ADAP receives rebates under the 340B Rebate option. I will not go into the 340B Rebate option in great detail.
The way a Medicaid or ADAP rebate works is that for a brand-name drug you get a minimum of 15.1 percent off the average manufacturer's price (AMP), which is a secret price. Though you cannot know what that price is, you get 15.1 percent off the AMP, or the difference between AMP and the best price that the pharmaceutical company is giving to a private sector payer, if it is larger. Plus, you get a supplemental rebate if the AMP price increases exceed the inflation rate as stated by the Consumer Price Index (CPI). Inflation is calculated back to the initial introduction of the drug, which is a little complicated, but it means typically when you are in a low-inflation period, such as we have been, the rebates get larger if the drug prices have increased more quickly than the rate of inflation. If you get a 5 percent price increase, what an ADAP or Medicaid is really going to see is a net cost creep of about 0.50 percent or so per quarter, and it keeps getting recalculated each quarter, keeping track with the CPI.
The rebate for generic drugs, of which there are no antiretroviral drugs in this case, is 11 percent of AMP; the 340B discount price is simply AMP minus that rebate amount.
Slide 2 is an attempt to make things a little simpler to comprehend. When you are direct purchasing, that price is going to be the bottom block of the figure. This is a hypothetical drug that costs US$100 at AWP. I say it is very hypothetical in the antiretroviral drug business because we would love to see a US$100 drug, and there is nothing anywhere close to that. The AWP is the retail, or list, price. The figure shows a discount. The pharmacy does not get the full amount as discounted. In this case, the example was discounted 10 percent. The next amount goes to the pharmacy and the wholesaler; it is their cost of doing business, for taking possession of the drug and moving it to the pharmacy; and to the pharmacy for handing it to the individual, and all of the counseling and handling costs involved in that. The next piece comes back as a rebate. But the amount of money going to the pharmaceutical company is really US$68 in this hypothetical situation. That is important, because we talk a lot about drug pricing but it is very convoluted, and the price that you see or that we are even able to talk about is not necessarily the price that anyone is paying. Joshua Cohen alluded to this issue in his presentation (see "Untangling the Economics of Drug Pricing" in this issue).
New York State's ADAP pays AWP minus 12 percent to the pharmacy. This is the same rate that Medicaid pays. If an ADAP were to purchase drugs directly, we could take out the cost of the rebate, the discount to the pharmacy, the pharmacy wholesaler amount, which is their cost of distribution, and their profit margin, plus the potential for additional price concessions. There is a difference between this price, which is closer to the net cost, and an AWP minus 12 percent price, which is what you would see in government data.
There are variations in price increases for individual drugs. You have got the slow steady creep of zidovudine (ZDV). It is the turtle in the race. The company that manufactures this drug takes regular price increases, lets inflation grow, and takes advantage of that CPI penalty or makes sure that there is always headroom so it can grow. When efavirenz (EFV) entered the market in 1998, we had a problem. It was priced substantially higher than other drugs. It set a new pricing point. The only drug that has broken from that trend is emtricitabine (FTC), for which we are seeing cost parity because it has essentially the same effectiveness as lamivudine (3TC).
Whenever we have a new drug breaking new pricing points, it has a negative impact for payers for the next round of drugs coming through the pipeline. Besides some of my other problems with ritonavir (RTV), it really affected the scale. Again, in a similar scenario, we had consistency on pricing for quite a while. Then enfuvirtide (ENF), admittedly a new class of drug, was introduced, but following that there was a new price point; and the price points are coming up further. For fosamprenavir (FPV), in fairness, a boosted dosage is about half the amount of an unboosted dosage, but we are still trying to see what the mix is going to be between boosted or unboosted dosages. The big issue is probably the substantial difference in many cases between the actual cost and the public price. There is a substantial difference between what it looks like we pay and what we actually do pay.
Slide 3 shows the average total monthly cost of antiretroviral drugs per user, and here we have a very clear trend. It is generally a result of more drugs being used in combination. Plus, there are newer drugs coming in, and those new drugs tend to be higher priced, despite the efforts we make to keep prices down. It is also due to regular price increases on the part of most pharmaceutical companies. It is getting more expensive to keep a person on antiretroviral therapy.
There is an ancillary cost as well with antiretroviral drugs, which is treating the toxicities and side effects. For example, in Slide 4 we took a basket of drugs that we assumed would be primarily used to treat the side effects and toxicity of antiretroviral therapy. You have a big jump in 2000 because we added new categories that we had not added in the past. By 2000, we realized that we were absent some classes of drugs that are needed to adequately care for someone with HIV in the longer term. Relative to the actual cost of antiretroviral drugs per month, it is not a large amount, but it is a cost that is growing and it is part of the cost of treating HIV/AIDS.
Looking at New York State's ADAP costs by quarter from 1996 through 2004 (Slide 5), the nucleoside reverse transcriptase inhibitors (NRTIs) are still the major cost component. The protease inhibitors (PIs) are still the second largest. Non-nucleoside reverse transcriptase inhibitors (NNRTIs) and the new category of fusion inhibitors are third and fourth, respectively, in terms of overall spending on antiretroviral drugs. All of the other drugs, and New York is lucky to have about 450 drugs other than the antiretroviral drugs on its formulary, make up a relatively small part, about 14 percent of total cost. Again, I blame RTV for messing up my charts on average price increases here. We went into cost containment mode the first quarter of 2003 while we tried to get some more money, and we were fortunate that we did, largely from the New York State government. But it really did not have much impact since we could not have any control over the antiretroviral drug end of things, so we had to take the money out of the other drugs.
Clearly we have a growth trend here that is somewhat overwhelming in a relatively closed system such as ADAP. New York State's ADAP, unlike most other ADAPs, also has several other components. ADAP-Plus pays for ambulatory care, medical visits, and lab tests. We have a home care component, and we have ADAP-Plus Insurance Continuation (APIC). We do try and keep people in the private sector by helping them pay for their insurance premiums whenever possible. We did get some savings on home care from 1996 onward when antiretroviral drugs came into play, and we saw a dramatic decrease in home care, but there was little decrease in costs because home care was a small fraction of overall cost. Ambulatory care has more than doubled -- it is about two and a half times what it was, but the bulk of that cost is drugs, and the bulk of drug cost is the antiretroviral drugs. So, in a system where we are looking at primarily outpatient care, all we see are increasing costs year after year, month after month.
In New York State, Medicaid pays for the inpatient cost for their patients, and generally for the ADAP patients as well, because by the time they go into inpatient status Medicaid will kick in for them. We can see some of that cost effectiveness that was discussed earlier. In Slide 6 we can see that total Medicaid cost for AIDS dropped from 1994 to 1997 and then started to level off until 2001 when it started to increase again; this is the last year for which I was able to get data. The one thing that I have to point out is that costs dropped from 1994 to 2001 by about US$200,000 a year. Medicaid's antiretroviral drug bill went up by about by about US$500 million in that same time period. Even though they were saving US$200 million in one place, they were still accruing about US$300 million more in HIV costs at this point, so it is not a net savings, it is a net loss that we are dealing with for government.
It is becoming a harder sell to keep HIV-specific and -related drugs exempt from cost containment measures. Fortunately, our trump card is the decrease in AIDS deaths in New York State from 1990 to 2002. That dramatic decrease in deaths that had started occurring in 1996 and 1997 has somewhat leveled out now. The result of that decrease, of course, is more people living with HIV -- they are living longer, better lives. What you are seeing is that because there are more people now living with HIV and requiring ongoing care and treatment, we are accumulating more and more costs.
So, what is the impact on government and patients? Increasing ADAP and Medicaid costs came at a particularly bad time over the last few years. Both the fiscal situation in this country, and the economy, have not been good. States are suffering a great deal, and it is being reflected in how states are dealing with their ADAPs and Medicaid programs. There is an indirect cost on government through rising health insurance premiums. Michael [Allerton] will deal with that in more detail, I am sure. But from a government perspective, we are seeing individuals losing their private insurance. They can no longer afford to pay those rising premiums, so they are turning to the state programs. There is also a rising cost to government for our own employee health costs, and the impact ultimately on the patients comes by way of the cost containment measures that are being imposed.
Medicaid is focused on generic substitution, preferred drug lists, prior authorization, and reduced pharmacy reimbursement rates. They are trying to squeeze the pharmacies and, ultimately, they are also trying to squeeze pharmaceutical companies with some supplemental rebates by using preferred drug lists. We are very fortunate in New York State; we have probably the most stable ADAP in the country. When we went into cost-containment mode in 2003, it was around mandatory generics, a limit on the number of refills before the person has to be seen again and issued a new script, prior authorization on selected drugs, and replacement of high-cost drugs with lower-cost alternatives. We reduced the amount of Ambien® people could get, and they screamed more about that than anything else. We were very fortunate that ours was more of a tightening rather than elimination. Other ADAPs are much worse off, with capped enrollments and waiting lists.
We are probably going to find that the January 2004 number of 791 people on ADAP waiting lists that José [M. Zuniga] referred to earlier is going to be the lowest number we will see for a long time. States are moving to reduce financial eligibility, imposing restrictive medical criteria, and reducing formularies. I think the most dramatic example of that is New Jersey, which had a full formulary in March 2004. On April 1, 2004, that state's ADAP formulary was downsized to only antiretroviral drugs. ... So they went from abundance down to a very minimal formulary for treating HIV. We are seeing monthly limits on the number of prescriptions allowed, or the cost per participant. That cost per participant is a particular problem when the drug prices go up, because the cost per person has to be calculated at the front end of what is being paid at the pharmacy, not at the back end of what is the net cost for the pharmacies. When you use cost containment like that, it is not really fair to the individual patient.
There are some major problems with current government pricing systems. We have a fragmented system with discrete pricing schedules for the various government entities. It is secretive pricing, and it is protected by law. State officials cannot give you the exact price that we are paying. We cannot even know sometimes whether the price that we are paying is the correct price. There are loopholes in the government pricing protections. The CPI penalty that comes into effect if drug prices increase faster than the rate of inflation goes away if there is a new formulation of an existing drug. If the new formulation gets a new code, it starts from scratch and we are only guaranteed that 15 percent rebate. So new formulations, while a really good thing from a clinical perspective, are not necessarily such a great deal for us from a payer perspective, and we find the pharmaceutical industry blocking any legislative attempts to reform pricing.
I am very glad to be here with you, and with IAPAC today, to discuss what I hope will be some ideas. There have been some efforts on the part of government and the community to partner around drug pricing. The Fair Pricing Coalition is a coalition of treatment advocates and government payers, which engages senior management of pharmaceutical companies in pricing discussions. The primary focus has been on initial pricing of new drugs and on requesting price freezes from industry. The ADAP Crisis Task Force, which was formed last year, represents 10 ADAPs or AIDS Center Directors, reflecting 70 percent of the buying power of ADAPs. We negotiated pricing concessions with all of the antiretroviral drug manufacturers. We calculated US$60 million to US$65 million nationally in savings for ADAPs in that first round of discussions. All ADAPs benefited equally, even though there were only 10 ADAP representatives in the room; one of our major points was that all ADAPs will get a share of the savings. We coordinate efforts with the Fair Pricing Coalition and other advocacy groups, and we intend to continue to expand our efforts to secure the best possible prices for ADAPs.
You have seen a lot of concise, scientific information about cost effectiveness and quality years of life. Unfortunately, when you are working in government, sometimes the people who are making decisions are only seeing my charts with the big growth lines and the increasing bottom line for government agencies, and that is a problem. It is really dependent upon physicians and treatment advocates to make sure decision-makers get the full picture, but you must realize that often they are looking from a very limited perspective, bottom line tax dollars. How much can we raise taxes and get away with it? How much do we have to cut in order to maintain government as it is? We must play a role in helping them to see the bigger picture.
Lanny Cross is Program Manager of New York State's AIDS Drug Assistance Program.
Back to the May 2004 Supplement issue of IAPAC Monthly.