Health Affairs, 24, no. 4 (2005): 1117-1127 doi: 10.1377/hlthaff.24.4.1117 © 2005 by Project HOPE	New Online   How Would Obama, McCain Cover The Uninsured?   Debating Cost Of Uninsured   Try Medicare-For-All   HA Blog Top 10

This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (8) Citing Articles via Google Scholar Google Scholar Articles by Mullins, C. D. Articles by Palumbo, F. B. Search for Related Content PubMed PubMed Citation Articles by Mullins, C. D. Articles by Palumbo, F. B. Related Collections Insurance Market Business Of Health Consumer Issues Insurance Coverage Pharmaceuticals Health Spending

DataWatch

Variability And Growth In Spending For Outpatient Specialty Pharmaceuticals

Abstract

 
Specialty pharmaceuticals are a unique group of drug agents used to treat complex clinical conditions. Many specialty pharmaceuticals are biological in nature and administered through injection or infusion. Tracking spending on these pharmaceuticals is complex, because these products may be processed as either medical or pharmacy claims. This benchmarking study of ten Blue Cross Blue Shield plans, representing almost eighteen million covered lives, documents large expenditures on select specialty pharmaceutical categories and much variation in spending across plans, age groups, and time. Our results underscore the need for insurers to scrutinize trends in specialty pharmaceutical spending and identify appropriate management strategies.

Explosive growth. Specialty pharmaceuticals have grown explosively in both numbers of agents and expenditures. In the United States in 2001, approximately $22 billion was spent on them.⁴ One year later, that figure had increased to $35 billion.⁵ This sector of the drug industry reports an average annual cost of approximately $71,000 per patient.⁶ The annual cost of specialty drugs is "growing two times faster than traditional ambulatory drug products."⁷ Pharmaceutical Research and Manufacturers of America’s (PhRMA’s) 2002 New Medicines in Biotechnology Survey found 371 "biotechnology medicines" in development that were either in human clinical trials or awaiting U.S. Food and Drug Administration (FDA) approval by 144 manufacturers and the National Cancer Institute.⁸ These drugs represent only a subset of specialty pharmaceuticals.

Many of these specialty medications provide highly sophisticated treatment for rare or chronic conditions, such as non-Hodgkin’s lymphoma, for which there previously were no viable pharmaceutical options.⁹ These products provide value, but with prices averaging more than $1,000 a month per patient—and rising at a higher rate than that for regular medications—the challenge lies in how to best manage the use of these drugs to ensure appropriate and affordable access.¹⁰

Tracking difficulties. Spending for specialty pharmacy is difficult for insurers to track.¹¹ Claims for these products may enter the system through multiple points: pharmacy submissions, billings from physicians’ offices, home care agencies, and outpatient facilities such as outpatient hospital clinics. The payment records for these products are stored on multiple data systems and are rarely tracked comprehensively. However, given the explosive growth in both the number of products available and spending for the products, good comparative information on spending is increasingly important. Many insurers are contemplating a variety of payment and distribution strategies, all of which will be difficult to evaluate in the absence of good trend information.

BCBS benchmarking study. As a response to this need, the Blue Cross and Blue Shield Association (BCBSA) worked with a consortium of BCBS plans to conduct a benchmarking study to identify plans’ levels of spending on expensive injectable drugs and specialty pharmaceuticals in the outpatient setting. The first aim was to determine the level of spending and variability across plans, insurance business types, and two age groups—under sixty-five versus sixty-five and older. The second aim was to explore the growth in spending across two years.

Study Data And Methods

Top Study Data And Methods Study Findings Policy Implications NOTES

 
Two panels from participating BCBS plans were used to select the specialty pharmaceuticals and data elements for this analysis. The first, the Medical Panel, was convened to review the drug classes and provide expert opinions on drug usage. Most of the Medical Panel members were pharmacists, although six were physicians. The second, the Technical Panel, consisted of data analysis and research-oriented staff, such as computer programmers, who had expertise in the layout of their own plan’s specific databases. The use of parallel panels was crucial because of the need to extract comparable data from different plans’ databases.

Selection of drug classes. Individual specialty pharmaceuticals and drug classes were selected based on the perceived likelihood that these agents would reflect the current high costs to the plans or were likely to exhibit higher usage in upcoming years. Furthermore, only drugs with an existing Healthcare Common Procedure Coding System (HCPCS) or Current Procedural Terminology (CPT) codes were included in the specialty pharmaceutical classes. The only exception was for unclassified therapeutic agents (J3490 and J9999), which included drugs that did not have their specific CPT or HCPCS codes.

We included only drugs that had both National Drug Codes (NDCs) and HCPCS/CPT codes. NDCs were used to identify specific drugs from the pharmacy claims, whereas CPT and HCPCS codes were used to identify drugs from medical claims. This allowed plans to examine their total drug spending regardless of the benefit design. In addition, unclassified therapeutic agents were included because of their high spending amounts, even though there was no corresponding NDC available. The final drug classes and specific drugs are listed in Exhibit 1.

Spending components. Expenditures included health plan payments, patients’ copayments and deductibles, and any coordination-of-benefits payments across multiple insurers. We compared data for two full years across plans. The first year consisted of claims data for 1 October 2001–30 September 2002 and paid through 31 December 2002; the second year included claims data for 1 October 2002–30 September 2003 and paid through 31 December 2003.¹³ Data from pharmacy and professional medical claims were included in the study. We excluded data on outpatient facility claims because some plans did not have procedure codes for them. This results in an underestimation of spending but greater comparability across plans. All plans provided data for 2002 and 2003 on drug spending, as well as on aggregate enrollment and the number of users of specialty pharmaceuticals, for each of the twenty categories identified in Exhibit 1.

The analysis compared three measures of utilization: spending per enrollee, rate of utilization per 10,000 enrollees, and spending per patient. Spending per enrollee provides an indication of how the drug category could affect insurance premiums. It also is the best way to assess the impact of total spending independent of changes in membership. Rate of utilization provided comparative information on how frequently the products were being used and growth in the rate of use. The study did not look at prices paid for individual drug agents. However, spending per patient does give an indication of how much the insurer spent each year for a patient who required products from one of the identified drug classes.

Insurance types. Each plan categorized each claim into one of three business types: managed care, preferred provider organization (PPO), and traditional. Managed care plans were defined as insurance products where an enrollee chooses a primary care provider upon enrollment. PPOs were defined as those not requiring a particular primary care provider but where the enrollee can select from a generally large panel of participating providers. Traditional plans were defined as those that reflected the more traditional fee-for-service model. Plans submitted either individual claims or aggregate spending data on each drug class for each business type and age group (under sixty-five versus sixty-five and older).

Study Findings

Top Study Data And Methods Study Findings Policy Implications NOTES

 
Data from ten BCBS plans, representing 17.8 million covered lives, were available for comparative analysis. Because of variability at the individual plan level in size and business-type mix, for analysis we included managed care and traditional data from eight plans and PPO data from nine plans. The breakdown in size and business-type mix also varies according to the stratified age categories—under sixty-five and sixty-five and older. Overall, in 2003 there were more than fifteen million enrollees in the first age group and more than two million in the second.

The ten BCBS plans reported $690,586,396 in spending in 2002 and $928,985,207 in 2003 for the products listed in Exhibit 1. The identified specialty pharmacy categories represented 13.9 percent of total pharmaceutical and specialty pharmaceutical spending for 2002, and 16.3 percent of total spending in 2003. Thus, from 2002 to 2003, annual spending for the identified specialty pharmacy categories increased 34.5 percent. This is accounted for by an increase of 22.9 percent in the total cost per enrollee and 9.4 percent in the number of enrollees. Growth in spending per patient similarly documents that the cost increase is the strongest factor in driving spending upward. There was a 20.0 percent increase in the cost per patient and a 12.1 percent increase in the use of products across the twenty specialty pharmaceutical categories. In 2003 the weighted average payment per enrollee was $41.11 for people under age sixty-five and $130.38 for people age sixty-five and older, up approximately one-fourth from the 2002 levels of $33.31 and $102.25, respectively. For the most part, payments per enrollee were highest for the traditional insurance business type, followed by managed care and PPO. One exception to this was the weighted average in the managed care group age sixty-five and older, which was higher than both traditional and PPO plans in 2003 and much higher than the amount reported in the prior year. Overall payment per enrollee per year for specialty pharmaceuticals in the younger age group increased 23.4 percent from 2002 to 2003, with individual plan changes ranging from a 10.0 percent decrease to an 814.7 percent increase. In the older age group, overall payment per enrollee per year increased 27.5 percent from 2002 to 2003, with individual plan changes ranging from a 37.0 percent decrease to a 198.7 percent increase.

Since the weighted averages give greater weight to larger plans, it also is informative to examine data on the median, minimum, and maximum plans by business type, age group, and year. For managed care and PPO in the younger age group, the median plan tracked closely with the weighted average. However, in that group the median plan was much higher for traditional insurance, thus demonstrating that plan size was not the sole determinant of payments. Interestingly, when we look at per enrollee median plan payments in the older age group, managed care and traditional more closely tracked with the weighted average. We observed enormous differences between the minimum and maximum plan payments, with more than a tenfold difference in the cost per enrollee for several specialty pharmaceutical categories.

High-spending categories: payment per patient. Among the twenty categories, pharmaceuticals with the highest per patient cost per year for 2003 were hemostatics (primarily used for the treatment of hemophilia) at $96,302, with an increase of 19.4 percent from 2002; platelet inhibitors for the treatment of pulmonary arterial hypertension (PAH) at $29,538, with an increase of 37.7 percent from 2002; and serums at $22,102, with an increase of 20.4 percent from 2002 (Exhibit 2). Many of the high-spending pharmaceuticals are used for chronic therapy. For example, the weighted average payment per patient taking hemostatics was by far the greatest among the specialty pharmaceutical categories. Platelet inhibitors, serums, and anti-neoplastic–monoclonal antibodies represented the second-, third-, and fourth-largest categories, respectively. However, to keep the matter in perspective, the sum total of these three latter categories was less than that of the hemostatics.

View larger version (29K): [in this window] [in a new window]   EXHIBIT 2 Overall Weighted Average Payment For Specialty Pharmaceuticals, Per Patient, 2002 And 2003

View larger version (30K): [in this window] [in a new window]   EXHIBIT 3 Overall Weighted Average Payment For Specialty Pharmaceuticals, Per Enrollee, 2002 And 2003

It is important to note the difference in absolute numbers when we examine the data per enrollee because large percentage increases sometimes reflect a low baseline rate. Payment per enrollee for hemostatics was $1.46 in 2002 and $1.81 in 2003. For 2003, blood modifiers for treating anemia represent the largest category per enrollee ($6.76 per enrollee in 2002 and $8.45 in 2003), while anti-arthritics, unclassified therapeutic agents (those without an HCPCS code) and anti-neoplastics–luteinizing hormone-releasing hormone (LHRH) agonists represent the second-, third-, and fourth-largest payment categories, respectively. Unclassified therapeutic agents present some analytic challenges. The amount of money spent per enrollee is much higher than that paid for most other categories, and yet there is incomplete information regarding which specialty pharmaceuticals these expenditures represent.

High-trend categories: payments per patient and per enrollee. Unclassified therapeutic agents accounted for some of the largest growth percentages across the two years. When we examine overall weighted average payment per patient, the data show a 49.5 percent increase from 2002 to 2003. Growth in payments per enrollee from 2002 to 2003 is almost identical at 50 percent (Exhibit 3). The other category with the largest change in payment per patient from 2002 to 2003 is platelet inhibitors–PAH, with a 37.7 percent increase over 2002 (Exhibit 2). The other categories with the greatest year-to-year change per enrollee are anti-neoplastics–enzyme inhibitor (60.1 percent), platelet inhibitors–PAH (53.2 percent), and antivirals (55.0 percent). Fertility treatment products decreased from 2002 to 2003 in both payment per patient and payment per enrollee. The only other decrease from 2002 to 2003 was seen with payment per enrollee for blood modifiers–other, which reflects only one drug, oprelvekin, for treatment of severe thrombocytopenia.

Increase in use. Increases in medical spending in general and specialty pharmaceutical spending in particular come from both increases in price and increases in the quantity used. Among the twenty specialty pharmaceutical categories, those with the highest usage among patients under age sixty-five were unclassified therapeutic agents, antiemetics, and viscosupplement, both in 2002 and 2003 (Exhibit 4). For the group age sixty-five and older, the categories with the greatest number of users were anti-neoplastics–LHRH agonists, blood modifiers–anemia, and viscosupplement. Fifteen of the twenty categories experienced increased usage among the younger age group, with a range of 1.7–34.9 percent. In particular, anti-neoplastics–enzyme inhibitor (43.1 percent), antivirals (respiratory syncytial virus, or RSV) (34.9 percent), and anti-arthritics (32.7 percent) demonstrated the highest growth. For the older age group, of the thirteen categories that experienced an increase in usage, antivirals (RSV) and fertility treatment increased 131.2 percent and 110.3 percent from 2002 to 2003, respectively, followed by anti-neoplastics–enzyme inhibitor (53.7 percent) and interferons–beta (21.6 percent). Note that the percentage increase in fertility treatments for the older age group is off a very small baseline measure, since use among the elderly is negligible.

Top Study Data And Methods Study Findings Policy Implications NOTES

The shift to specialty pharmaceuticals often results in higher spending on pharmaceuticals because these agents tend to be more costly than older therapies. At the same time, these agents may be cost-effective and may occasionally produce cost savings. Payments per enrollee for the unclassified therapeutic agents were the highest category among enrollees under age sixty-five and the fifth-highest category among enrollees age sixty-five and older. This dynamic category is essentially the "catchall" category for unclassified drugs (J3490) and unclassified anti-neoplastic drugs (J9999) that do not have their own assigned procedure codes. Typically, drugs that are new to the market are included in this category.

Two major obstacles make it difficult for plans to monitor their spending for these new and expensive drugs. First, when a service provider submits a claim for any HCPCS code for unclassified drugs, the provider must submit the appropriate explanation of drug name and dosage under item 19 of the CMS Form 1500 and the corresponding electronic format. This information will be converted into a nonstandard text format in the claims, which makes it very difficult to identify the specific drug that was administered. Second, for a drug to be assigned its own HCPCS code requires approval from the CMS. This national coverage determination process is a lengthy five-step process, which includes regulatory approval, determination of the statutory eligibility to be deemed a benefit, determination of the extent of coverage (local versus national), assignment of an HCPCS code, and distribution of the decision and implementation instructions. It can take as long as nine months from the time a coverage request is submitted to the CMS to the time of approval.¹⁵ Until a specific procedure code is established for a drug, it is very difficult for plans to monitor the true spending. Consequently, it will take a prolonged period of time before plans can assess the trend of drug spending and then develop strategies to ensure appropriate use of the drug.

Specialty pharmaceuticals represent a major cost to insurers, and, as this benchmarking study has shown, spending on these agents is growing rapidly. Therefore, greater evaluation of their clinical effectiveness, appropriateness of use, medical cost offsets, and cost-effectiveness is warranted.

Editor's Notes

 
C. Daniel Mullins (dmullins{at}rx.umaryland.edu) is professor and chair, Pharmaceutical Health Services Research (PHSR) Department, at the University of Maryland School of Pharmacy in Baltimore. Andrea DeVries is head, Health Services Research, at Highmark Inc. in Pittsburgh, Pennsylvania. Van Doren Hsu is director of pharmaceutical research computing, Fanlung Meng is information systems engineer, and Frank Palumbo is director of the Center on Drugs and Public Policy, all in the PHSR Department.

The authors gratefully acknowledge the assistance of Mary Madison, Nadine Caputo, Lisa Blatt, the Pharmaceutical Research Computing staff at the University of Maryland School of Pharmacy, and the Blue Cross and Blue Shield Foundation on Health Care staff, as well as the plans that participated in the Outpatient Specialty Pharmaceuticals Benchmarking Study: Blue Cross and Blue Shield of Illinois, Blue Cross and Blue Shield of Hawaii, Blue Cross and Blue Shield of Michigan and Blue Care Network of Michigan, Blue Cross and Blue Shield of North Carolina, Blue Cross and Blue Shield of Rhode Island, CareFirst BlueCross BlueShield (Maryland), Highmark Inc. (Pennsylvania), Horizon Blue Cross and Blue Shield of New Jersey, and Wellmark Blue Cross and Blue Shield of Iowa (and South Dakota). The authors received funding for this research from the Blue Cross and Blue Shield Foundation on Health Care.

NOTES

Top Study Data And Methods Study Findings Policy Implications NOTES

 

1. D. Willcutts, "Specialty Pharmaceuticals: Developing a Management Plan," Managed Care Quarterly 10, no. 4 (2002): 26–29.[Medline]
2. B.L. Hesselgrave, "Helping to Manage the High Cost of Rare Diseases," Managed Care Quarterly 11, no. 1 (2003): 1–6.
3. D. Willcutts, "Developing a Management Plan," Managed Care Quarterly 11, no. 1 (2003): 7–10.[Medline]
4. National Institute for Health Care Management Research and Educational Foundation, Prescription Drug Expenditures in 2001: Another Year of Escalating Costs, 6 May 2002, www.nihcm.org/spending2001.pdf (6 December 2004).
5. M. Edlin, "Specialty Pharmacies Provide Pipeline for Low-Volume, High-Cost Biotech Drugs," Managed Healthcare Executive (1 June 2003): 36–37.
6. Advance PCS, Benefits Barometer, 2003, www.advancepcs.com/images/uploads/57.pdf (14 April 2005).
7. Hesselgrove, "Helping to Manage the High Cost of Rare Diseases."
8. A. Humphreys and C. Boersig, "Twelfth Annual Report, Top 100 Biotechnology Companies, the Ups and Downs of Biotechnology," Med Ad News 22, no. 1 (2003): 40–45.
9. P.C. Nagle, T.F. Lugo, and C.A. Nicita, "Defining and Characterizing the Late-Stage Biopharmaceutical Pipeline," American Journal of Managed Care 9, no. 6 Supp. (2003): S124–S135.[ISI][Medline]
10. R.J. Lipsy et al., "Anticipating the Future: How the Emergence of Innovative Biologic Agents Impacts Benefit Design, Utilization, and Provider Relations," Journal of Managed Care Pharmacy 10, no. 3 (2004): S4–S9.
11. M. Zitter, "Defining the Great Debate," American Journal of Managed Care 10, no. 8 Supp. (2004): S231–S239.[Medline]
12. Under the Social Security Act, Section 1883(t)(6), a transitional payment is a temporary additional payment for certain innovative devices, drugs, and biologicals.
13. A ninety-day period for claims payment is considered standard within the health insurance industry, with 95 percent or more of claims paid within this period.
14. Nagle et al., "Defining and Characterizing."
15. Lipsy et al., "Anticipating the Future."

                      What's this?

This article has been cited by other articles:

				C. D. Mullins, D. C. Lavallee, F. G. Pradel, A. R. DeVries, and N. Caputo Health Plans' Strategies For Managing Outpatient Specialty Pharmaceuticals Health Aff., September 1, 2006; 25(5): 1332 - 1339. [Abstract] [Full Text] [PDF]

				P. M. Danzon and M. F. Furukawa Prices And Availability Of Biopharmaceuticals: An International Comparison Health Aff., September 1, 2006; 25(5): 1353 - 1362. [Abstract] [Full Text] [PDF]

				C. Borger, S. Smith, C. Truffer, S. Keehan, A. Sisko, J. Poisal, and M. K. Clemens Health Spending Projections Through 2015: Changes On The Horizon Health Aff., March 1, 2006; 25(2): w61 - w73. [Abstract] [Full Text] [PDF]