Tracking Health Care Costs: Trends Slow in First Half of 2003

Data Bulletin No. 26
December 2003
Bradley C. Strunk, Paul B. Ginsburg

Health care spending growth per privately insured American slowed in the first half of 2003, increasing 8.5 percent, a significant drop from the 10 percent increase in the second half of 2002 (see Table 1).

While the cost trend remains extremely high in relation to the GDP trend, the 1.5 percentage point decline in spending growth in the first half of 2003 was the largest since the early 1990s. The slowing of the overall cost trend reflected slower growth in all four categories of health care spending—inpatient hospital care, outpatient care, prescription drugs and physician services.

Prescription drug spending growth slowed the most, rising only 8.5 percent in the first half of 2003, or nearly 5 percentage points less than the 13.4 percent increase in the second half of 2002. Moreover, the first half of 2003 represented the first time since the mid-1990s that the drug trend did not grow at double-digit rates.

Spending on inpatient hospital care grew 7.6 percent during the first half of 2003, down from the 8.3 percent increase in the last six months of 2002. Spending on outpatient care increased 12.9 percent in the first half of 2003, down from 14.1 percent in the second half of 2002. Outpatient care remains the fastest-growing category of health care spending. Spending on physician services increased 6.1 percent during the first half of 2003 and was again the slowest-growing category of health care spending.

TABLE 1: Per Capita Growtha in Health Care Spending and Gross Domestic Product (GDP), 1998-2003
 
Spending on Type of Health Care Service
 
All Servicesb
Hospital Inpatientb
Hospital Outpatientb
Physician
Rx Drugs
GDP
Jan.-June '98
4.9%
-0.5%
8.7%
4.0%
13.1%
4.4%
July-Dec. '98
6.6
2.3
8.7
5.5
15.0
4.3
Jan.-June '99
8.1
2.9
11.5
6.4
17.2
4.2
July-Dec. '99
6.8
1.8
10.5
3.7
19.5
4.6
Jan.-June '00
7.2
3.3
8.3
6.0
15.3
5.4
July-Dec. '00
8.5
5.1
7.7
1.6
11.0
4.4
Jan.-June '01
10.0
8.1
15.3
6.2
15.3
2.1
July-Dec. '01
10.3
10.1
14.7
7.1
12.4
1.2
Jan.-June '02
9.7
10.3
13.5
5.9
13.0
2.1
July-Dec. '02
10.0
8.3
14.1
7.1
13.4
3.2
Jan.-June '03
8.5
7.6
12.9
6.1
8.5
2.9
a Each row indicates the percentage change for the period indicated compared with the corresponding months in the previous year.

b Trend estimates have been adjusted to remove the effect of changes in hospital prices for Medicaid patients.

Notes: GDP is in nominal dollars.

Sources: Health care spending data are the Milliman USA Health Cost Index ($0 deductible); GDP is from the U.S. Department of Commerce, Bureau of Economic Analysis

Back to Top

Hospital Prices Up, Use Slows

After rising steadily but relatively slowly between 1997 and the first half of 2002 (see Table 2), the hospital price trend accelerated sharply in the last half of 2002 (6.2%) and the first half of 2003 (8.0%). Recent evidence from Center for Studying Health System Change site visits indicates hospitals continue to have formidable negotiating leverage in gaining large payment rate increases from health plans.1

Over the past few years, sharp hospital compensation cost increases—in response to a worker shortage—have created cost pressures for hospitals. For example, the hourly cost of compensating nurses at private hospitals grew by 8.8 percent during 2002, which was 3.5 percentage points greater than the increase in 2001 and 6.6 percentage points greater than the average increase from 1995 to 2000. Growth in compensation costs for nurses slowed somewhat during the first six months of 2003 but remained high by historical standards, suggesting hospitals may continue to face this cost pressure.

In contrast to the accelerating hospital price trend, the hospital utilization trend, which is calculated as a residual, slowed markedly, growing just 2.3 percent in the first half of 2003, down from 5 percent in the last half of 2002. The lower hospital utilization trend likely reflects the effect of increased patient cost sharing for hospital services and the completion of the transition to more loosely managed care.

Back to Top


Table 2: Hospital Spending Trend,a by Price and Quantity, 1998-2003
 
Change per Capita
  Spending on Hospital Servicesb Hospital Prices Quantityc
Jan.-June '98
3.7%
2.0%
1.7%
July-Dec. '98
5.4
1.9
3.4
Jan.-June '99
7.1
2.4
4.5
July-Dec. '99
6.1
2.6
3.4
Jan.-June '00
5.8
2.8
3.0
July-Dec. '00
8.5
3.9
4.4
Jan.-June '01
11.8
4.0
7.6
July-Dec. '01
12.5
3.3
9.0
Jan.-June '02
12.0
4.3
7.4
July-Dec. '02
11.4
6.2
5.0
Jan.-June '03
10.5
8.0
2.3
a Each row indicates the percentage change for the period indicated compared with the corresponding months in the previous year.

b Trend estimates have been adjusted to remove the effect of changes in hospital prices for Medicaid patients.

c Calculated as the residual of hospital spending and hospital price trends.

Sources: Data on hospital spending are from the Milliman USA Health Cost Index ($0 deductible) and include both hospital inpatient and outpatient services; data on hospital prices are from the Bureau of Labor Statistics' "Other payors" Producer Price Index series for general and surgical hospitals

Outlook

The slowing of underlying cost trends is likely to bring an end, perhaps in 2004, to the long period of accelerating employersponsored health insurance premium trends, which reached a 13-year peak in 2003 at 13.9 percent.2 This development will provide some relief to private and public payers and consumers.

However, cost and premium trends are still likely to remain well in excess of trends in GDP for the foreseeable future. While increased patient cost sharing is probably an important factor in the slowing of cost trends, few expect this tool to substantially lower cost trends over the long term. Meanwhile, new tools, such as disease management and tieredprovider networks, remain unproven in the marketplace from a cost control standpoint. Without more effective cost-control measures, the rising cost of health insurance will make coverage unaffordable to more Americans, resulting in an increase in the uninsured.

Back to Top


This Data Bulletin is based on data from the Milliman USA Health Cost Index (HCI) ($0 deductible),which is designed to forecast claims trends faced by private insurers and the U.S. Bureau of Labor Statistics'National Compensation Survey to track hourly compensation costs for nurses and Producer Price Index for general medical and surgical hospitals to track hospital prices. The HCI classifies spending on services performed in freestanding facilities in its hospital outpatient category, which is consistent with how insurers classify such services. Due to data limitations, the HCI includes spending for Medicaid and uninsured patients,which can cause HCI trends to differ from privately insured trends. The authors have adjusted the HCI estimates to remove the effect of distinct Medicaid hospital price trends. As with most economic data, the HCI is subject to periodic retroactive revisions.


Notes

1. Lesser, Cara S., and Paul B. Ginsburg, Health Care Cost and Access Problems Intensify: Initial Findings from HSC's Recent Site Visits, Issue Brief No. 63, Center for Studying Health System Change, Washington, D.C. (May 2003).
2. Gabel, Jon, et al., "Health Benefits in 2003: Premiums Reach Thirteen-Year High as Employers Adopt New Forms of Cost Sharing," Health Affairs, Vol. 22, No. 5 (September/October 2003).

Data Bulletins are published by the Center for Studying Health System Change (HSC)
President: Paul B. Ginsburg
Vice President: Len M. Nichols