This Article Abstract Submit a response to this article Services E-mail this article to a friend Alert me to new issues of the journal

H E A L T H  T R A C K I N G
T R E N D S  

21 June 2005  
  

Tracking Health Care Costs: 
Declining Growth Trend 
Pauses In 2004

The deceleration in health spending trends has flattened out
but at a higher level than before, which alarms some observers.

By Bradley C. Strunk, Paul B. Ginsburg, and John P. Cookson

ABSTRACT:

Health care spending increased 8.2 percent in 2004. This was virtually unchanged from 2003, which suggests that health care cost trends have stabilized. Hospital spending grew 10.1 percent in 2004, also virtually unchanged from 2003, reflecting a small increase in the hospital utilization trend and a small decline in hospital price inflation. Meanwhile, growth in prescription drug spending continued to fall as a result of slower growth in prices. Growth in health insurance premiums slowed again in 2005, likely reflecting earlier years’ slowing in cost trends and signaling that a turn in the insurance underwriting cycle might be under way.

The trend in private health insurance premiums slowed in 2004 after a long period of acceleration culminating in a peak increase of 13.9 percent in 2003. Nevertheless, premium growth continues to outpace growth in the economy and workers’ incomes by a wide margin, making health care benefits increasingly unaffordable for employers and employees alike.¹ The high rate of growth in premiums is primarily the result of rapid growth in the health spending that underlies private health insurance premiums—that is, the cost of services typically covered by private health insurance. Underlying cost trends are the dominant long-term determinant of premium trends.²

One year ago we reported that cost trends underlying private health insurance slowed after reaching their highest levels in more than a decade in 2001 and 2002.³ This paper updates that analysis and reports that the deceleration that began in 2003 has ended as quickly as it began. Growth in health spending in 2004 was virtually unchanged compared to 2003, reflecting stable trends in each of the major health service categories except prescription drugs, which decelerated for the fifth year in a row.

Meanwhile, health insurance premium growth slowed further in 2005, probably reflecting the lagged relationship between underlying cost trends and premium trends. This is welcome news for those who pay for coverage. Employers also may be shifting costs to their employees in the form of higher patient cost sharing at a slower rate. However, the stabilization of the cost trend at a relatively high rate may foreshadow a similar development for premiums in the near future.

Study Data And Methods

Cost trend data. We used the Milliman Inc. Health Cost Index (HCI) to examine recent trends in health care spending underlying private health insurance premiums. Milliman constructs this index from both publicly available and proprietary data on provider revenues (a proxy for spending on services) gathered through various provider surveys. The index is designed to reflect private insurers’ claims expenses for a typical policy.⁴ Although Milliman removes spending by Medicare from these series, its inability to remove spending by Medicaid and uninsured patients is an important limitation of the HCI’s ability to track the spending trends underlying private insurance. Nevertheless, a past comparison of the HCI and the National Health Accounts (NHA) compiled by the Centers for Medicare and Medicaid Services (CMS) indicated that the HCI is a good measure of private health care costs.⁵ We used the HCI instead of the NHA because the HCI is available with a shorter time lag.

Milliman made several important changes and revisions to the HCI in 2004 that resulted in changes to the trend estimates in this report as compared with past reports. In the past, the HCI has tracked changes in four major categories: inpatient and outpatient hospital care, physician care, and prescription drugs. In 2004 Milliman added a fifth category, “All Other,” which primarily includes spending on home health care, ambulance services, and some other types of ambulatory services. Prior to 2004, these services were included in the outpatient hospital category. The outpatient hospital category in the Milliman data continues to include freestanding facilities, such as surgery and imaging centers.⁶

Also, Milliman made several revisions to the way in which the overall spending index is calculated, most notably by revising the estimates of the percentage of spending attributable to each major service category. These revisions led to changes in the rate of growth in health care spending in each year of the study period (that is, an average of 1.8 percentage points between 1993 and 2002 as compared with Exhibit 1 of our 2004 paper), but the changes in growth rates from one year to the next did not change much.⁷ Since the services in the new “All Other” category have a lower trend than those still included in the hospital outpatient category, the creation of the new category increased the trend in the latter.

To gain insight into the factors driving growth in spending on hospital care, we broke the spending trend down into its price and quantity components. We used the “all other payers” series of the Bureau of Labor Statistics (BLS) Producer Price Index (PPI) for general medical and surgical hospitals to measure changes in hospital prices for privately insured patients. This series (hereafter referred to as the “hospital PPI”) reflects negotiated payments rather than billed charges.⁸

Changes in hospital quantity (use, length-of-stay, and resource intensity per case) were calculated indirectly as the residual of the HCI for hospital services (inpatient and outpatient combined) and the hospital PPI. This quantity index is subject to the same limitation as the HCI—that is, it includes changes in prices and use for Medicaid and uninsured patients. However, a past analysis has shown that the effects of this limitation are small and would not likely change our overall conclusions.⁹

We decomposed the trend in spending on prescription drugs in similar fashion. We used the Consumer Price Index (CPI) for prescription drugs and medical supplies to measure changes in drug prices. Like the hospital PPI, the prescription drug CPI reflects actual transaction prices. It accounts for price changes for the existing roster of prescription drugs, as well as changes in prices paid by consumers from the increased use of generic substitutes or lower-price brand-name drugs.¹⁰ We also calculated the residual of the HCI and CPI for prescription drugs. This residual captures the change in the total number of prescriptions written per person (that is, drug utilization), which itself reflects changes in (1) the proportion of the population that receives at least one prescription during the year, and (2) the average number of prescriptions among all people that receive at least one (that is, the intensity). It also reflects changes in the therapeutic mix of drugs prescribed and the introduction of new brand-name drugs. When drugs move to over-the-counter (OTC) status, the HCI, and hence our residual, reflects the reduced drug spending but does not have an offset for what is spent over the counter.

Finally, we used data on payroll costs for private (nongovernment) hospitals to elucidate changes in hospitals’ largest operating cost factor. These data, compiled monthly by the BLS through its Current Employment Statistics survey, are useful for their reliability and very short time lag. We report BLS payroll data per capita because this makes them directly comparable to the HCI and to data on premiums (what is charged to cover an individual or family) and therefore the most relevant measure for policymakers.

Premiums and cost sharing. We examined findings from two employer surveys, the Towers Perrin 2005 Health Care Cost Survey and the National Business Group on Health (NBGH) and Watson Wyatt Worldwide Employer Survey of Trends in the Health Care Marketplace, to glean insights into health insurance premium trends for 2005.¹¹ Both of these surveys are limited by the fact that they rely on convenience samples. The Towers Perrin survey included responses from 385 employers representing approximately 5.4 million covered employees, retirees, and dependents. Participating employers were asked to report their 2005 per capita premium costs for insured health plans and premium equivalents for self-insured plans. The NBGH/Watson Wyatt survey included responses from 555 employers with 1,000 or more full-time employees, that collectively provide benefits to nearly ten million employees. The Towers Perrin survey also provides an estimate of the role played by increased patient cost sharing and reduced benefits—referred to as “benefit buy-down”—in the premium trends.¹²

Another important limitation of these surveys for our purposes is that the samples do not include small employers. For this reason, we focused the most attention on how the 2005 increase compared with the 2004 increase. Since premium increases for small employers have tended to outpace those for large employers in recent years, the estimates cited here may underestimate the increase across employers of all sizes.¹³

Health Care Spending Trends

Total health spending per privately insured person increased 8.2 percent in 2004 (Exhibit 1), a rate of growth that was virtually unchanged from 2003. This trend stabilization is a departure from the slowdown in the two years prior to 2003, when spending growth declined almost three percentage points from a peak increase of 11.3 percent in 2001. Health spending outpaced overall economic growth by a wide margin (2.6 percentage points) again in 2004, despite a 5.6 percent nominal increase for the U.S. economy, as measured by per capita gross domestic product (GDP).

Hospital spending. The rate of growth in spending on hospital inpatient services per privately insured person remained largely unchanged in 2004 (Exhibit 1). Although these growth rates represent a slowdown relative to 2001, when the inpatient trend peaked at 8.6 percent, they remain very high relative to the mid-1990s, when spending on inpatient care actually declined from one year to the next. The trend for spending on hospital outpatient care also was largely unchanged in 2004. Outpatient hospital care was by far the fastest-growing category of health care spending, despite the fact that its trend has fallen from a peak of 14.5 percent in 2001. Taken together, growth in spending on inpatient and outpatient hospital care accounted for 54 percent of the total increase in health care spending in 2004, which is much greater than the share of spending attributable to hospital care.

The stabilization of spending trends for hospital care conceals divergent trends for hospital use and unit prices (Exhibit 2). The trend for hospital use (as measured by our residual hospital quantity index) remained low, increasing 2.9 percent in 2004. Nevertheless, this increase was 1.2 percentage points larger than the increase in 2003, which may be attributable in part to the recent economic recovery driving greater demand for hospital services, as well as to transitory factors affecting trends. At the same time, growth in unit prices for hospital services fell a percentage point from 2003, the year in which hospital price inflation peaked.

The slowdown in the hospital price trend is particularly notable because it represents the first such slowdown in seven years (Exhibit 2). Hospitals have sought higher and higher payment rate increases from health plans for several years, in part to make up for agreeing to severely discounted rates in the mid 1990s in expectation of growing health maintenance organization (HMO) enrollment. It is likely, however, that the large increases hospitals received, because of their growing leverage in contract negotiations, have now restored margins on privately insured patients to targeted levels for at least some hospitals.¹⁴ There is no evidence to suggest that hospitals’ negotiating leverage has waned, but we may be seeing hospitals tempering their demands after a successful period of catch-up pricing. Indeed, we might have seen a larger decline in price trends if not for declines in margins for publicly insured patients, which might be leading to higher targets for margins on privately insured patients.¹⁵ Medicare margins dropped between 2002 and 2003, and although we are not aware of data on Medicaid margins, it would be surprising if they were not falling as a result of state budget constraints.¹⁶

The slowdown in the hospital price trend also could reflect a continuing slowdown in hospitals’ most important operating expense: payroll for hospital workers. A shortage of hospital workers, particularly nurses, has driven up wage rates in recent years, but that shortage appears to have passed its short-term peak.¹⁷ In 2004 the average hourly wage for hospital workers grew 4.8 percent, which was 0.6 percentage points higher than in 2003 but still 1.3 percentage points lower than in 2001, when wage rates and perhaps the nurse shortage peaked (Exhibit 3). At the same time, total hours worked by hospital personnel grew 1.2 percent, down from a 2.0 percent increase in 2003. The trend in total payroll expenses per capita, which reflects the combined effect of wage trends and growth in hours worked, fell to 6.0 percent in 2004 from its high of 8.1 percent in 2001. A slowdown in payroll expenses is probably just starting to show up in the hospital price trend, because hospitals would have had to negotiate their payment contracts with health plans before recognizing developments with respect to the nursing shortage.

Prescription drugs. For the fifth year in a row, the drug spending trend decelerated in 2004 (Exhibit 1). Spending on prescription drugs per privately insured person grew 7.2 percent in 2004 compared with 8.9 percent in 2003. Moreover, this increase was less than half the 1999 increase of 18.1 percent. Growth in drug spending accounted for 21 percent of the total increase in health care spending, well below its contribution in the late 1990s.

The continuing slowdown in 2004 is largely the result of slower growth in drug prices rather than a change in use (Exhibit 4). Prescription drug prices increased just 3.3 percent in 2004 compared with 5.2 percent in both 2002 and 2003.¹⁸ This likely reflects, in part, the movement toward three-tier copayments, which have now become common in health benefit offerings, as well as the continuing growth in those copayments across all of the three tiers.¹⁹ Three-tier designs encourage greater use of generic drugs, which have become increasingly prevalent in recent years and tend to be priced far below their brand-name equivalents. Also, differences in copayments for preferred versus nonpreferred brand-name drugs help create price competition among manufacturers that could be helping hold down price increases.

In contrast to the slowdown in drug price growth, drug usage trends changed little between 2003 and 2004. This follows a major slowdown in trend between 2002 and 2003, when growth in use fell from 7.5 percent to 3.5 percent. That slowdown was likely attributable in large part to the reclassification of some widely used drugs, such as the nonsedating antihistamine Claritin and the proton-pump inhibitor Prilosec, to OTC status. Other factors that may be contributing to continued slow growth in drug usage are higher copayments and new information about the negative side effects of hormone replacement therapy.

Physician care and other services. Spending on physician care per privately insured person increased 6.4 percent in 2004—identical to the rate of growth in 2003 (Exhibit 1). A decomposition of physician spending comparable to that in Exhibit 2 for hospital spending shows a small decline in the rate of growth of physician utilization and a small increase in growth of prices for physician services. Meanwhile, spending on other types of services, most notably home health care and ambulance services, grew 6.0 percent and was the slowest-growing health spending category. Growth in spending on physician care and all other services accounted for 24 percent and 2 percent of the total increase in health care spending, respectively.

Health Insurance Premium Trends

Findings from the Towers Perrin and NBGH/Watson Wyatt employer surveys suggest that trends in premiums for employment-based insurance continue to slow in 2005 but remain high. The Towers Perrin survey reports that premiums for active employees increased 8 percent on average this year, down from the 12 percent increase in 2004. Similarly, the NBGH/Watson Wyatt survey found that large employers expect a median premium increase of 10 percent this year, whereas last year’s respondents predicted that premiums would increase by a median rate of 12 percent in 2004. Net of benefit buy-downs, however, the decline in premiums was even greater, as buy-downs were smaller in 2005 than they were in 2004. When benefit structure does not change over time, premium trends tend to be higher than the trend in underlying costs because the insurer is paying a higher percentage of claims costs.

Because premium trends tend to lag underlying cost trends by one to two years, this slowdown likely reflects the slowdown in underlying health care cost trends that occurred between 2002 and 2003. Also, a portion of this decline may signal an impending turn in the health insurance underwriting cycle—the insurance industry’s interdependent pattern of profitability and pricing—toward its so-called soft phase. During the soft phase, insurers engage in vigorous price competition, sometimes at the expense of profitability, to gain market share. Some Wall Street analysts are predicting a deterioration of medical loss ratios in 2005, which lends credence to the possibility of an impending turn in the underwriting cycle.²⁰

Implications For Consumers

In recent years employers have moved to increase patient cost sharing, such as deductibles, copayments, and coinsurance, as a way to cope with double-digit premium increases. This trend appears to have continued in 2005 for the fourth year in a row, although perhaps to a lesser degree than in the recent past. According to a Wall Street analysis of premium pricing data compiled by Hewitt Associates, employers bought down benefits by 0.7 percent in 2005.²¹ Respondents to the Towers Perrin employer survey, meanwhile, reported a somewhat higher average reduction in benefit levels (2 percent). Nevertheless, both of these estimates appear to indicate a deceleration relative to the 2002–2004 period, when buy-downs ranged from 2 percent to 4 percent, and may signal either that employers have begun to reach what they consider the limits of acceptable patient cost sharing or that they feel less pressure to shift costs to employees in light of the improving economy and recent declines in the premium trend.²² At the same time, employers continue to refrain from raising employees’ share of premiums. In 2005 employees are paying 18 percent of the cost of single coverage and 22 percent of the cost of family coverage—levels that have been unchanged for several years.²³

Outlook For The Future

The question of when purchasers of health insurance might expect to see some relief from rapidly rising health care costs and insurance premiums has become a perennial one, and the results of this study suggest that the answer remains elusive. Unfortunately, the good news of one year ago that health care cost trends turned downward has now been replaced by the bad news that trends stabilized in 2004 at a relatively high rate of growth. Moreover, uncertainty remains high about where trends might be headed next. Nevertheless, a number of forces operating in the health care system today bear watching as potential drivers of cost trends in the near future.

A number of evolving developments with respect to prescription drugs could have a material impact on the drug spending trend in the near future. First, the withdrawal of the blockbuster COX-2 inhibitor Vioxx in the fall of 2004, together with related Food and Drug Administration (FDA) actions that have occurred since then—including the removal of another COX-2 inhibitor, Bextra, from the market and the requirement that a strict warning be added to the only remaining COX-2 inhibitor, Celebrex—will likely place strong downward pressure on drug usage in 2005. Meanwhile, a number of other important blockbuster drugs, such as the antidepressant Zoloft, are scheduled to lose patent protection in the next couple of years, which will hold down drug price increases as generics are introduced. In contrast, spending on specialty pharmaceutical products, such as injectibles, is rising rapidly and placing upward pressure on drug spending trends.²⁴ Specialty drugs have accounted for a minor portion of total drug spending in the past, but this share could increase quickly if current trends prevail into the future.

Several additional forces on the horizon could affect hospital spending. Hospitals might slow the degree to which they are increasing margins for privately insured patients, but possible substantial cuts in Medicare payment rates could prevent this from happening. A high degree of capital spending by hospitals, emphasizing expansion of capacity to provide specialty services, could lead to higher spending, either through mechanisms of supply creating demand or through hospitals’ anticipation of demands for additional services that come from advancing medical technology. In addition, the 2005 Community Tracking Study (CTS) site visits show that hospitals are facing increasing competition from new and expanding physician-owned facilities (both freestanding and in their offices). This could contribute to growth in spending for the reasons described above, as well as through the effects of physician self-referral.

One important question that applies across all spending categories is the degree to which further increases in cost sharing will drive down the use of health care. Common assumptions are that demand for health care services is inelastic (that is, people are insensitive to price differences) and static, which suggests that cost sharing ultimately will not have a large impact on the trend of health care use.²⁵ However, to the degree that incentives are structured to encourage use of lower-cost providers or treatment alternatives, the impact could be larger and could affect trends as well as rates of spending. A second issue that cuts across all categories of spending is the degree to which greater use of information technology (IT) in health care could lower costs in the future. Policy analysts tend to be divided on whether or not IT will have a major impact on costs, but regardless of the outcome of that debate, it seems unlikely that investments in IT will have any material effect in the near term.

Models used to forecast future health care spending trends usually incorporate a strong but lagged relationship between general economic growth and health care spending.²⁶ The years of recession and weak recovery are likely to continue to limit trends in health care spending, although if the strong economic growth in 2004 continues, it could press spending trends upward a few years out. If the recent commodity-based inflation continues, this could add to spending growth as well, although not necessarily to the extent that inflation based more on rising wage rates would do.

The forces brewing in today’s health care system are varied, and it does not appear as though they will be aligned in any one direction to drive cost trends either higher or lower. It seems particularly unlikely that cost trends will resume a sustained decline in the next year or two, especially since the most important long-term driver of health care spending—advances in medical technology and its ready acceptance into mainstream medical practice—shows no sign of slowing. Indeed, a “magic bullet” to contain costs continues to elude us, and we are left to confront the sobering knowledge that more health care cost growth in excess of growth in workers’ incomes will lead to more uninsured people.²⁷ Although there is no clear consensus among policymakers and major stakeholders in this country about how best to control costs, the imperative to continue that debate and act in some fashion remains as high as ever.

The authors gratefully acknowledge the Robert Wood Johnson Foundation for its financial support, as well as the valuable comments of two anonymous reviewers.

NOTES

1. J. Gabel et al., “Health Benefits in 2004: Four Years of Double-Digit Premium Increases Take Their Toll on Coverage,” Health Affairs 23, no. 5 (2004): 200–209.
2. We use the terms “cost” and “spending” interchangeably. Conceptually, the primary interest is in costs, which reflect the resources devoted to health care that are not available to produce other goods and services. Practically, most available data, including those used in this paper, reflect spending, or what is paid for health services by those who purchase them (or received by providers of health services).
3. B.C. Strunk and P.B. Ginsburg, “Tracking Health Care Costs: Trends Turn Downward in 2003,” Health Affairs, 9 June 2004, content.healthaffairs.org/cgi/content/abstract/hlthaff.w4.354 (19 May 2005).
4. The index that Milliman provides to its clients is intended to assist insurers in forecasting their claims payments and comparing them with those of others. It simulates trends in claims for a “standard” private health insurance policy with a $250 deductible and various other types of cost sharing for physician office visits, hospital care, and prescription drugs. The trend in such an index would slightly overstate the actual trend in spending because the standard policy would pay for a slightly higher proportion of expenditures each year. To avoid this problem, we use a version of the index that reflects a hypothetical policy with no deductible but with the implicit utilization levels of the plan with copayments and deductibles. Since the market has been reducing benefits over the past few years by increasing copayments and deductibles, which have a reducing effect on utilization, the index would produce somewhat lower trends than plans with no actual change in benefits are likely to experience.
5. B.C. Strunk, P.B. Ginsburg, and J.R. Gabel, “Tracking Health Care Costs: Growth Accelerates Again in 2001,” Health Affairs, 25 September 2002, content.healthaffairs.org/cgi/content/abstract/hlthaff.w2.299 (30 March 2004).
6. This differs from the National Health Accounts, which include these freestanding facilities in the physician category rather than in the hospital outpatient category.
7. We did not include the 2003 estimate in this comparison because it, unlike the estimates for 1993–2002, had not yet undergone routine revisions. The HCI, like other economic series, is periodically revised as better data become available.
8. For more information about the hospital PPI’s methodology, see B. Catron and B. Murphy, “Hospital Price Inflation: What Does the New PPI Tell Us?” Monthly Labor Review 119, no. 7 (1996): 24–31. We considered the hospital portion of the CPI as an alternative to the hospital PPI, but the hospital CPI is less useful for our purposes because it uses billed charges for hospitals not responding to its survey. The PPI, in contrast, drops such nonresponders.
9. Strunk and Ginsburg, “Tracking Health Care Costs: Trends Turn Downward.”
10. For more information, see Bureau of Labor Statistics, “Measuring Price Change for Medical Care in the CPI,” 16 October 2005, stats.bls.gov/cpi/cpifact4.htm (17 May 2005).
11. Towers Perrin HR Services, 2005 Health Care Cost Survey (New York: Towers Perrin, 2005); and National Business Group on Health and Watson Wyatt Worldwide, Managing Health Care Costs in a New Era: Tenth Annual National Business Group on Health/Watson Wyatt Survey Report 2005 (Washington: Watson Wyatt Worldwide, 2005).
12. A buy-down of, for example, 2 percent means that insurance premiums would have increased 2 percent more than they did if employers had made no changes to their benefit structures.
13. For comparisons of premium increase for small versus large employers, see G. Claxton et al., Employer Health Benefits: 2004 Annual Survey (Menlo Park, Calif.: Henry J. Kaiser Family Foundation, September 2004); and previous versions of this survey.
14. For information about hospitals’ leverage in contract negotiations, see J. White, R.E. Hurley, and B.C. Strunk, “Getting Along or Going Along? Health Plan–Provider Contract Showdowns Subside,” Issue Brief no. 74 (Washington: Center for Studying Health System Change, January 2004).
15. P.B. Ginsburg, “Can Hospitals and Physicians Shift the Effects of Cuts in Medicare Reimbursement to Private Payers?” Health Affairs, 8 October 2003, content.healthaffairs.org/cgi/content/abstract/hlthaff.w3.472 (12 May 2005).
16. For data on Medicare margins for hospitals, see Medicare Payment Advisory Commission, Report to Congress: Medicare Payment Policy (Washington: MedPAC, March 2005).
17. P.I. Buerhaus, D.O. Staiger, and D.I. Auerbach, “New Signs of a Strengthening U.S. Nurse Labor Market?” Health Affairs, 17 November 2004, content.healthaffairs.org/cgi/content/abstract/hlthaff.w4.526 (12 May 2005).
18. Based on discussions with the BLS, we have decided to report the 2003 prescription drug price increase as 5.2 percent rather than the published estimate of 3.1 percent. We did this because the prescription drug CPI continued in 2003 to include the nonsedating antihistamine Claritin and the proton-pump inhibitor Prilosec, even though these drugs were reclassified to over-the-counter status in December 2002 and September 2003, respectively. According to the BLS, the drug CPI would have grown 2.1 percentage points more in 2003 had these two drugs been removed from the sample at the time of their respective reclassifications. Bureau of Labor Statistics, CPI Services Division, personal communication, 12 May 2005; estimates are calculations using the research index calculator.
19. Gabel et al., “Health Benefits in 2004: Four Years of Double-Digit Premium Increases.”
20. C. Arnold et al., Managed Care Industry Overview—Hewitt Final Round: Rates Slip, Buydowns Dip (New York: Morgan Stanley Equity Research, 20 October 2004).
21. D. Simpson and M. Jaffe, Managed Care Industry Report—Hewitt Pricing Data: In Line with Expectations (New York: Merrill Lynch, 20 October 2004). Employers that did not buy down benefits in 2005 are included with those that did in the calculation of this estimate, making it a true national average of benefit buy-down. Among only those employers with buy-down greater than zero, the average buy-down was 2 percent.
22. For more information on buy-downs in 2002–2004, see Strunk and Ginsburg, “Tracking Health Care Costs: Trends Turn Downward.”
23. The Towers Perrin survey includes responses from mostly large employers, and large employers usually pay a larger portion of the cost of family coverage than small employers pay. This explains why the Towers Perrin estimate of the employee share of family coverage is lower than those of other employer surveys that include all employers.
24. B. Motheral et al., Express Scripts 2003 Drug Trend Report (St. Louis: Express Scripts, June 2004).
25. Other researchers have made similar arguments; for example, see J.P. Newhouse, “Consumer-Directed Health Plans and the RAND Health Insurance Experiment,” Health Affairs 23, no. 6 (2004): 107–113.
26. For more information, see J. Cookson, “Outlook for Health Care Trends,” 11 January 2001, sihp.brandeis.edu/council/pubs/spending/Cooksonpaper.htm (17 May 2005).
27. T. Gilmer and R. Kronick, “It’s the Premiums, Stupid: Projections of the Uninsured through 2013,” Health Affairs, 5 April 2005, content.healthaffairs.org/cgi/content/abstract/hlthaff.w5.143 (20 April 2005).

Bradley Strunk (BStrunk{at}hschange.org) is a health researcher at the Center for Studying Health System Change in Washington, D.C. Paul Ginsburg is the center's president. John Cookson is a principal with the consulting firm Milliman Inc. in Wayne, Pennsylvania.

DOI: 10.1377/hlthaff.w5.286
©2005 Project HOPE–The People-to-People Health Foundation, Inc.