Abstract
- Issue: Between 2013 and 2017, spending on Medicare Part D catastrophic coverage more than doubled, reaching over $59 billion in 2017. In the catastrophic phase, beneficiaries are responsible for 5 percent coinsurance, which can result in high out-of-pocket costs.
- Goals: To inform redesign of the Part D catastrophic benefit by providing data on: beneficiary characteristics in the catastrophic phase; factors driving entry into catastrophic coverage; and patterns of spending, specialty drug use, and utilization among these beneficiaries.
- Methods: Analysis of Medicare Part D claims and beneficiary data from 2013 to 2017.
- Key Findings: Among Part D beneficiaries not receiving Medicare low-income subsidies, those who entered catastrophic coverage in 2017 averaged $3,218 in total annual out-of-pocket spending, compared to $486 among enrollees who did not enter catastrophic coverage that year. Beneficiaries who entered catastrophic coverage had more chronic conditions, but the biggest difference was specialty drug use. Close to 60 percent of catastrophic coverage enrollees took a specialty drug compared to 11 percent of beneficiaries who did not enter this coverage phase.
- Conclusions: To reduce the high drug cost burden for Medicare Part D enrollees in catastrophic coverage, policy options should focus on the use and prices of specialty drugs.
Introduction
In 2017, Medicare spent $156.5 billion on prescription drugs through Part D. Over time, an increasing portion of this spending has been on catastrophic coverage. From 2013 to 2017, spending in the catastrophic phase increased more than twice as fast as overall Part D spending, reaching $59 billion in 2017, or about 40 percent of Part D spending.1 Spending on high-priced specialty drugs is responsible for most of the increase, making up two-thirds of Part D catastrophic spending in 2015, compared to one-third in 2009.2
Part D is administered through private prescription drug plans, which have the ability to negotiate drug prices with manufacturers and pharmacies and employ formularies, tiered copayments, and other utilization management tools (within Medicare rules). Once the catastrophic portion of the benefit is reached, the plan pays 15 percent of the cost, Medicare pays 80 percent, and the beneficiary pays the remaining 5 percent. Because Medicare covers most of the price of the drug, Part D plans have little incentive to negotiate aggressively for high-price specialty drugs. MedPAC has recommended changing the spending distribution to give more spending risk to the plans.3
Under the standard Part D benefit, beneficiaries were responsible for the following in 2017: a deductible, then 25 percent coinsurance up to an initial coverage limit, followed by a coverage gap (“donut hole”) in which beneficiaries paid 40 percent (brand) or 50 percent (generic) of the drug price until their total out-of-pocket spending reached a given limit ($4,950 in 2017). After beneficiaries exit the coverage gap, they enter catastrophic coverage.4 For many Medicare enrollees, the 5 percent cost-sharing requirement in the catastrophic coverage phase results in high spending burdens that can severely limit access to specialty pharmaceuticals, unless beneficiaries are eligible for patient assistance programs.5
There is little information on the characteristics of beneficiaries most affected by the out-of-pocket cost-sharing burden in Part D catastrophic coverage or what drives them into this phase. When Part D was established, policymakers expected that beneficiaries with multiple chronic conditions who took multiple drugs would be the primary entrants to catastrophic coverage given long-term use of multiple prescription drugs among this group. Increasingly, however, the growth of specialty drugs means that beneficiary spending can reach high levels because of a single drug.6 No research has yet determined how this trend and other market changes have impacted beneficiary entrance to and experience in Part D catastrophic coverage.
Improving our understanding of Part D catastrophic coverage is an important first step in designing policies to help ensure that beneficiaries are able to afford drugs through Medicare Part D. We need to understand who enters the catastrophic coverage phase of Medicare Part D, how much they spend before and after they enter this phase, how they differ from beneficiaries who do not enter catastrophic coverage, and what factors push beneficiaries into this final phase.
We used enrollment and claims data for Medicare Part D for 2013–2017 to assess the characteristics of beneficiaries in Part D catastrophic coverage; drivers of catastrophic entry; and patterns of enrollee spending, specialty drug use, and overall utilization in the catastrophic phase. For each year, we limited the sample to those in stand-alone prescription drug plans who remained in the same plan for the full year and did not die during the year. The results were analyzed separately for enrollees who received the Part D low-income subsidy, since their cost-sharing, usage, and expenditures may differ in systematic ways from other enrollees. We also examined prescription drug use among the subgroup of patients with multiple chronic conditions (based on current treatment for a chronic condition) to be able to separate out the effects of taking a specialty drug from having poor health. For further detail on the study methods, see “How We Conducted This Study.”
Findings
Spending in Catastrophic Coverage
Total and out-of-pocket drug spending during catastrophic coverage have increased over time. Among those not receiving the low-income subsidy, total per beneficiary spending in catastrophic coverage increased from $12,373 in 2013 to $22,031 in 2017, and average per beneficiary out-of-pocket spending in the catastrophic phase rose from an average of just over $900 in 2013 to $1,372 in 2017.
Unsurprisingly, increased spending in the catastrophic phase resulted in average annual drug costs that were 13 times higher for enrollees who entered catastrophic coverage compared to those who did not ($28,771 versus $2,183). As shown in the Appendix, average out-of-pocket costs were seven times higher among enrollees who entered catastrophic coverage compared to those who did not ($3,218 versus $486). Total spending among catastrophic entrants also grew much more over the five-year period: 53 percent compared to just 3 percent among those who did not enter the catastrophic phase.
Characteristics of Beneficiaries in Catastrophic Coverage
In 2017, Part D enrollees who entered catastrophic coverage and did not receive low-income subsidies had similar demographic characteristics — age sex, race — as enrollees who did not enter this coverage phase (see Appendix). However, those in catastrophic coverage had more chronic conditions (8.8 reported chronic conditions, on average, compared to 6.6 among those who never entered catastrophic coverage).
Enrollees in catastrophic coverage were also prescribed more drugs (an average of 20 drugs versus 11 drugs). Generic drugs represented a smaller proportion of total drugs among the catastrophic enrollee population (68%, 16.1 drugs on average) than the population not in catastrophic coverage (90%, 10.1 drugs on average). Rates of generic usage remained relatively consistent for all enrollees over the 2013–2017 time period.
The primary factor driving higher out-of-pocket costs among the catastrophic coverage population was use of specialty drugs. Among enrollees not receiving the low-income subsidy, close to 60 percent of those who entered catastrophic coverage took a specialty drug, compared to just 11 percent of enrollees who did not enter catastrophic coverage.
Entry into Catastrophic Coverage
The catastrophic benefit was initially designed primarily to protect beneficiaries with multiple chronic conditions who consumed multiple different drugs over the course of the year. Now, beneficiaries taking a specialty drug are more likely to enter catastrophic coverage than those with multiple (three or more) chronic conditions. Specialty drugs typically treat complex or chronic diseases and are identified by the Centers for Medicare and Medicaid Services (CMS) based on cost. Part D plans may put drugs with a high monthly cost (greater than $600 from 2013–2016 and greater than $670 in 2017) on a specialty tier of their formularies.7 In this study, we used these Part D monthly cost limits to identify specialty drugs. In general, these drugs have very high price tags; for example, the average retail price of Harvoni®, used to treat Hepatitis C, was $31,050 per prescription in 2015, and the average price of Revlimid®, used to treat anemia, multiple myeloma, and lymphoma, was $10,130.8
We found that specialty drug use was a bigger driver of catastrophic coverage entry than the presence of multiple chronic conditions (Exhibit 1). Nearly 25 percent of those without multiple chronic conditions taking a specialty drug entered catastrophic coverage, compared to only 4.8 percent of those with multiple chronic conditions who were not taking a specialty drug. In other words, controlling for the presence of multiple chronic conditions, those taking a specialty drug were about five times more likely to enter catastrophic coverage than those not taking such a drug.
Taking a specialty drug was also associated with higher out-of-pocket costs than having multiple chronic conditions. In 2017, beneficiaries with multiple chronic conditions and specialty drug use had the highest average out-of-pocket drug costs ($796) followed by enrollees who took specialty drugs but did not have multiple chronic conditions ($474). Beneficiaries who did not take specialty drugs had lower average out-of-pocket costs ($405 among those with multiple chronic conditions and $168 among those without multiple chronic conditions). This spending pattern suggests that specialty drug use is a more significant driver of out-of-pocket spending and catastrophic coverage entrance than the presence of multiple chronic conditions.
Beneficiaries in Catastrophic Coverage for Multiple Years
Many beneficiaries who entered catastrophic coverage did so year after year. Among the 597,155 enrollees not receiving the low-income subsidy who entered the catastrophic phase in 2017, 13 percent had catastrophic spending in all five years (2013–2017). Exhibit 2 shows the number of unique Part D enrollees in catastrophic coverage for one, two, three, and four years of our five-year sample. There were few consistently significant differences in age, sex, or race among beneficiaries who were in the catastrophic coverage all five years versus one or two years.
For those in catastrophic coverage for only one year, the two drugs with the largest total spending were Revlimid, which treats certain cancers, and Harvoni, a curative treatment for hepatitis C. The top two drugs for those in catastrophic coverage for all five years were Copaxone®, used to treat multiple sclerosis, and Abilify®, a psychotropic drug. These patterns suggest that beneficiaries who enter catastrophic coverage for many years are taking a different mix of specialty drugs than beneficiaries taking one expensive drug who enter catastrophic coverage for a single year. The proportion of people taking a specialty drug went up noticeably between 2013 and 2017 among enrollees in catastrophic coverage for multiple years. For example, among beneficiaries in catastrophic coverage in all five years, 51 percent took a specialty drug in their first year of catastrophic coverage compared to 61 percent in their most recent year of coverage.
Out-of-Pocket Spending in Catastrophic Coverage
Exhibit 3 shows the characteristics of beneficiaries without low-income subsidies who entered catastrophic coverage during our sample period. We compared various characteristics one month, six months, and 12 months before and after entry into catastrophic coverage.
Average total drug spending rose in the month following catastrophic coverage entry (from $1,804 in the month before entry to $3,080 in the month after entry). In this one-month period, spending was highly concentrated on the most expensive drugs (73% in the month following entry). However, this distribution decreased over time. Six months after entry, 62 percent of total spending was on the most expensive drug. Overall, average out-of-pocket spending dropped following entry into catastrophic coverage (from $530 in the month prior to entry to $203 in the month post-entry). This pattern is consistent with lower cost-sharing in the catastrophic coverage phase compared to the coverage gap period. The mean number of prescription fills and the percentage of generic fills were relatively consistent before and after entry.
Among beneficiaries who did not receive low-income subsidies, total out-of-pocket spending among those in catastrophic coverage was more than six times higher than those who did not enter catastrophic coverage (Exhibit 4). Among those who entered catastrophic coverage, out-of-pocket spending was highest in the coverage gap; however, spending was also high during the catastrophic phase, and the amount of catastrophic spending grew higher between 2013 and 2016. Total out-of-pocket spending in this population increased over the 2013–2016 period, driven by growth in catastrophic coverage spending. There was a small decrease in out-of-pocket spending in 2017.
Discussion and Policy Implications
Our analysis suggests that, in recent years, taking a specialty drug is a driving factor for why beneficiaries not on low-income subsidies reach the catastrophic phase. This was not the expectation when the Medicare Part D legislation was passed in 2003. The original belief was that beneficiaries with multiple chronic conditions taking multiple different drugs would comprise most of the beneficiaries reaching catastrophic coverage.
Beneficiaries entering the catastrophic benefit phase tend to be sicker and take more drugs, especially specialty drugs, relative to enrollees who do not reach this coverage level. For nearly all Medicare enrollees, rates of specialty drug use are going up over time, but this trend is especially pronounced for those who enter catastrophic coverage. The findings suggest that use of specialty drugs is the primary driver of entry into catastrophic coverage relative to the presence of multiple chronic conditions. Among those with multiple chronic conditions, close to 35 percent of those using specialty drugs entered catastrophic coverage in 2017, compared to 5 percent of those with no specialty drug use.
The financial burdens documented in this brief suggest that affordability is likely to be an issue for many of the sickest Medicare beneficiaries, especially those entering the catastrophic phase each year for multiple years. The recent legislation passed in the U.S. House of Representatives includes a cap on out-of-pocket costs for Part D enrollees, which will reduce the financial burden for individuals with especially high spending.9 This policy would relieve some of the cost-sharing burden on enrollees but is unlikely to result in lower drug prices and could even encourage drug manufacturers to increase prices given that the beneficiary would not have any cost-sharing in the catastrophic phase.
Related policies, such as the MedPAC proposal to adjust the burden of cost-sharing in the catastrophic period, could help lower prices. MedPAC recommends that beneficiaries pay nothing, Part D plans pay 80 percent, and Medicare pays 20 percent,10 which could help to strengthen plans’ negotiating leverage. The recently passed House bill also targets high prices by allowing CMS to negotiate prices for certain high-cost drugs with no generic competition and international prices used as a target price ceiling. Strengthening these negotiation efforts could play an important role in lowering list prices if Medicare has the necessary leverage. Without lowering list prices, the Medicare program will continue to shoulder the costs of expensive specialty drugs.
Conclusion
The Medicare Part D benefit structure has not been updated to reflect the growing dominance and cost of specialty drugs. Our analysis shows that specialty drugs are a major contributor to Part D spending and beneficiary out-of-pocket costs. Going forward, it will be critical to reduce enrollee burden and take broad actions to lower prices to ensure that drugs are affordable for Part D enrollees.
How We Conducted This Study
We identified Medicare Part D beneficiaries who entered catastrophic coverage using enrollment and claims data for Medicare Part D for 2013–2017. The sample consists of all prescription drug claims for a 20 percent sample of the Medicare population, which forms a nationally representative panel. The numbers in the brief have been scaled to be representative of all Medicare Part D beneficiaries in stand-alone prescription drug plans who had continuous coverage and did not die.
The claims data include detailed information on prescription drug characteristics, such as the total cost of the drug, the out-of-pocket amount paid by the beneficiary, and whether the drug was generic or brand name. The data also include personal characteristics of the beneficiaries, such as the prescription drug plan type, number of chronic conditions, portion of the benefit phase an individual is in, whether an individual receives a low-income subsidy, and basic demographic characteristics (such as age, gender, and race).
Because of the large sample size, there was sufficient power to analyze subgroups of the larger population. The study was able to investigate patterns in drug utilization by benefit phase of a beneficiary and receipt of low-income subsidies.
For each year included in the analysis, we limited the sample to those in stand-alone prescription drug plans who remained in the same plan for the full year and did not die during the year. We analyzed their demographics, prescription drug use, and drug-related expenditures. Then we compared these characteristics between beneficiaries who did and did not enter catastrophic coverage in the same year. In addition, the results were analyzed separately for enrollees who received the Part D low-income subsidy since their cost-sharing, usage, and expenditures may differ in systematic ways from enrollees who did not receive the low-income subsidy.
Prescription drug use among the subgroup of patients with multiple chronic conditions (based on current treatment for a chronic condition) was analyzed to separate out the effects of a specialty drug from having poor health. Chronic conditions were identified based on the 27 condition-specific variables available in the Centers for Medicare and Medicaid Services’ Master Beneficiary Summary File Chronic Conditions Data Warehouse Conditions segment. Finally, we compared drug utilization and spending for individuals at one month, six months, and 12 months before and after entry into catastrophic coverage during our sample period. For example, we assessed patterns of drug use and spending in the 12 months before and 12 months after catastrophic coverage entry for those who entered catastrophic coverage between January 2013 and December 2016.