Health Affairs, 24, no. 6 (2005): 1629-1639 doi: 10.1377/hlthaff.24.6.1629 © 2005 by Project HOPE	New Online   McAllen, TX & Beyond: An Expert Roundtable   Geography & Reform   Medicaid or Insurance Exchange?   Siren Song of New GME   Public Plan Option: Pro & Con

This Article Abstract Figures Only Reprint (PDF) Technical Supplement Submit a response to this article Alert me when this article is cited Alert me when Comments 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 (3) Citing Articles via Google Scholar Google Scholar Articles by Kahn, J. G. Articles by Gans, D. N. Search for Related Content PubMed PubMed Citation Articles by Kahn, J. G. Articles by Gans, D. N. Related Collections Hospitals Legal/Regulatory Issues Business Of Health Physicians State/Local Issues Health Spending State/Local Issues - California

DataWatch

The Cost Of Health Insurance Administration In California: Estimates For Insurers, Physicians, And Hospitals

Abstract

 
Administrative costs account for 25 percent of health care spending, but little is known about the portion attributable to billing and insurance-related (BIR) functions. We estimated BIR for hospital and physician care in California. Data for physician practices came from a mail survey and interviews; for hospitals, from regulatory reporting; and for private insurers, from a consulting company. Private insurers spend 9.9 percent of revenue on administration and 8 percent on BIR. Physician offices spend 27 percent and 14 percent, and hospitals, 21 percent and 7–11 percent, respectively. Overall, BIR represents 20–22 percent of privately insured spending in California acute care settings.

The reasons why this is the case are less clear. The hypothesis suggested by Woolhandler and colleagues, and supported by common sense, is that the complexities of a highly fragmented, multiple-payer system account for the "excess" administration. However, no direct data confirm this hypothesis. Of note, the high U.S. level relative to Canada could be accounted for in part by the greater complexities of running very large hospitals and medical groups, by greater scrutiny from public- and private-sector regulatory bodies, by greater effort devoted to utilization management and quality improvement, by a greater need to document activities in a litigious environment, or by U.S.-Canadian differences in the style and purpose of management and administration. To inform discussions of reform-related cost savings, it is valuable to document the portion of administrative costs attributable to the U.S. system of paying providers.

Problems of data interpretation present a major challenge for understanding administrative costs. Previous researchers have emphasized the limitations of any single-index measure of administrative cost (such as medical loss ratios), which reveal little of component costs, are defined inconsistently, and could be misleading if examined in individual health care sectors, because of complex health insurance arrangements.³ Instead, analysts have proposed a more nuanced scrutiny of administrative costs, including detailed classification that reflects functions, examination of how costs vary by administrative obligations, and a focus on the costs (and benefits) of specific activities—with the ultimate goal of understanding the overall implications of alternatives for administering the health system.⁴

Although administrative activities are sometimes thought of as "waste," some administrative effort is required and desirable in a well-functioning system. Hospitals are complex organizations, and administrative effort is needed to use inputs efficiently and produce good outcomes. As physician practice moves toward larger medical groups, administrative effort is required to assure that the groups function efficiently. Administrative activities here include the work of the office manager, the receptionist, the billing staff, the information technology experts, and other personnel not directly contributing to the hands-on care of patients.

The current emphasis on improving quality and reducing errors further highlights the desirability of some administrative effort.⁵ Similarly, public and private payers’ desire for greater accountability from the health care system creates demands for administrative activities. As payers and patients increasingly demand information on outcomes and expect efforts to improve quality, administrative effort is required. The information revolution arguably increases providers’ capabilities and improves the ability to monitor, measure, and improve health system performance, but it also requires administrative effort.

Yet administrative activities are not all created equal. For this paper, we divide administrative activities into those that are billing or insurance-related (BIR) and all others. The primary purpose of BIR activities is to move money from payer to provider in accordance with agreed-upon rules. We attempt to separate BIR functions from those whose primary purpose is general management or those primarily directed at measuring or improving quality. For some functions, this distinction is easier to make in theory than in practice. For example, to the extent that provider contracting at an insurance company is an effort to identify and contract with high-quality providers, it is part of quality improvement; to the extent that it is directed at reaching agreement on payment for services, it is BIR.

In this paper we estimate the fraction of health care spending for hospital and physician care in California that is devoted to BIR activities. Because we do not construct similar estimates for Canada and do not provide any detail on the non-BIR portion of administrative activities, we cannot fully answer the question of why U.S. administrative costs are so much higher than Canada’s. However, we begin to shed light on the contents of the black box of U.S. administrative costs.

Study Data And Methods

Top Study Data And Methods Study Results Discussion NOTES

 
We estimated the percentage of BIR costs in three settings: private insurers, physician offices, and hospitals. To accomplish this, we abstracted administrative cost elements from existing data sets, calculated overhead attributable to these costs, and estimated the percentage of each cost element that is BIR.⁶ We also describe how we estimated total percent BIR in California acute care funded through private insurance, and we examine the sensitivity of overall findings to uncertainty in particular BIR estimates. Our analysis best reflects administrative costs in 2000, based on the timing of input data.

Private insurers. We used data collected by Milliman USA during 1996–2001 at 73 insurers (7 in California) for 129 insurance plans (63 commercial, 43 Medicare, and 23 Medicaid). Median plan membership was 250,000. The data were provided to us as median cost per member per month for each plan type and cost element. The median premium per member per month, used as the denominator in calculating percent BIR, was $177 for commercial plans, $153 for Medicaid, and $650 for Medicare.

We divided private insurers’ costs into thirteen categories (Exhibit 1). Milliman collected data on the salary costs of employees involved in each category and added 150 percent for estimated overhead (employee benefits, information technology, and facilities). External brokers’ fees and profits are excluded.

Physician offices. For overall cost by category, we used 2000 data from the Medical Group Management Association (MGMA) annual survey of its member physician group practices.⁷ We used data from the U.S. Western region rather than California alone because of sample-size considerations. The valid response rate was 23 percent. We grouped responses into three categories: multispecialty (52 groups, 2,682 physicians), single-specialty primary care (27 groups, 221 physicians), and single-specialty surgery (15 groups, 84 physicians). Means are weighted by number of physicians. We estimated physician time spent on administration from a national survey conducted by the American Medical Association (AMA).⁸

We divided physician practices’ administrative costs into thirteen categories (Exhibit 2). Overhead was proportionally allocated to these categories.

Across the eleven practices, variation in the BIR percentage was lowest for the four administrative cost categories representing 70 percent of nonprovider BIR costs (business office, managed care administration, information services, and medical reception) and was higher for the four categories representing 30 percent of nonprovider BIR costs (Exhibit 2). This suggests consistency across practices in our respondents’ judgments about the percentage of activity devoted to BIR in the areas responsible for most of BIR.

Nonphysician clinical staff contributed 7.5 percent of their time to BIR (1.6 percent of practice revenue). For physicians, we did not ask systematically about BIR but instead relied on survey results showing that physicians spend five hours per week (8 percent of their time) on administration and other nonclinical tasks.⁹ Based on the breakdown of tasks and our estimates of BIR, we estimated that 4.9 percent of physician time (2.1 percent of revenue) is dedicated to BIR.¹⁰

Hospitals. We used fiscal year 1999 hospital financial data reported to the California Office of Statewide Health Planning and Development (OSHPD) for 392 acute general and children’s hospitals. Almost all hospitals reported data for each of the required data elements, with the exception of Kaiser Foundation hospitals, for which we were able to include Northern California hospitals in broad administrative and fiscal totals and some specific categories.

We divided administrative costs into twenty-one categories (Exhibit 3). Overhead was allocated to these categories.

Standardization across settings. To facilitate comparison across settings, we grouped spending into four categories. "Broad administrative" represents nonspecific administration categories, which can be large. The estimated proportion attributable to BIR varies by setting. "Claims billing or payment" represents activities explicitly linked to movement of funds related to insurance and has high BIR. "Other specific administrative" represents explicitly defined categories, with BIR level appropriate to the category. "Major clinical element" represents activities with clear clinical components, with BIR depending on the category and setting.

Total across settings. We calculated a total BIR across settings. In particular, we estimated the contribution of BIR administrative costs to acute medical care services (hospital and physician) funded through private insurance in California. We calculated BIR for these services by first applying the BIR for commercial plans to all spending via this mechanism, and then applying the BIR for multispecialty groups and the BIR range for hospitals to the portion of funds used in each clinical setting. For this overall estimate of BIR for privately insured care, we calculated the results with several measures of insurer operational profit (range 1.84–2.72 percent).¹¹

To simplify the analysis, we restricted our estimates of BIR to insurers, hospitals, and physicians. Although a more complete analysis would include estimated BIR for pharmacy and other insured services, our analysis captures the major types of acute care spending and 80 percent of all privately insured health spending.

Study Results

Top Study Data And Methods Study Results Discussion NOTES

 
Private insurers. For private insurers in the Milliman sample, administrative costs represent an estimated 9.9 percent of commercial plan premiums, 11.4 percent of Medicaid premiums, and 4.5 percent of Medicare premiums (Exhibit 1). BIR administration represents an estimated 8.4 percent, 9.4 percent, and 3.8 percent, respectively. BIR equals 85 percent of administrative costs for commercial and Medicare plans, and 81 percent for Medicaid plans. The largest BIR categories are claims, sales and marketing, finance and underwriting, and information systems, each accounting for 1.1–1.6 percent of premiums.

Physician offices. Medical groups in the MGMA sample report spending 20–27 percent of revenue on administration (Exhibit 2). BIR administration, calculated using BIR percentages from our California physician office interviews, represents 13.9 percent of total revenues for multispecialty, 14.5 percent for single-specialty primary care, and 12.4 percent for single-specalty surgery, respectively. BIR equals 52–61 percent of administrative costs for the three practice types. The largest BIR categories are the business office (3.5 percent for multispecialty groups), provider time (3.7 percent), information technology (1.7 percent), medical receptionists (1.1 percent), and administrative supplies and services (1.2 percent).

Hospitals. California acute care hospitals report spending 20.9 percent of revenue on administration (Exhibit 3). BIR administration represents an estimated 6.6–10.8 percent, depending on the BIR percentage value used for hospital administration and other administration. We used a wide range of values for these two categories to reflect our uncertainty about the appropriate allocation to BIR. Thus, BIR equals an estimated 31–51 percent of hospital administrative costs. The largest BIR categories are hospital administration (2.2–6.5 percent), patient accounting (1.6 percent), and credit and collections (1.0 percent).

Comparison across settings. Exhibit 4 summarizes results by setting, using the predominant modes for physician groups (multispecialty) and insurers (commercial plans). Administrative costs are highest in physicians groups. BIR costs are also highest in physician groups, with insurers and hospitals having similar BIR levels. BIR administrative costs as a percentage of all administrative costs are highest for insurers, which is consistent with their insurance function.

Because of uncertainty in our BIR estimates, we examined the sensitivity of these results to variation in our setting-specific estimates. We found that each ten-percentage-point-change in the BIR estimate in a specific setting results in a one-percentage-point-change in the overall BIR estimate. For example, we estimated that 52 percent of physician administrative costs are BIR, and if the true value is 42 percent, then BIR range drops from 19.7–21.8 percent to 18.7–20.8 percent.

We classified all health care spending for privately insured California hospital and physician care into four categories (Exhibit 5). For the private health care sector, BIR represents about one-fifth of costs, divided nearly equally between insurers and providers.

View larger version (14K): [in this window] [in a new window]   EXHIBIT 5 Allocation Of Spending For Hospital And Physician Care Paid Through Private Insurers

Top Study Data And Methods Study Results Discussion NOTES

Study limitations. Our analysis has important limitations. First, we assembled a variety of diverse data sources, none of them ideally suited for estimating BIR as a percentage of total costs. For hospitals and insurers, we relied on two authors’ judgment to estimate the portion of effort in each area that is devoted to BIR. These estimates certainly contain some error, although we believe that the authors’ competing leanings on level of BIR generated estimates that are not much different from those that many others would make. We were also reassured that many administrative tasks were reasonably assigned 100 percent or 0 percent BIR; they are, by definition, entirely or not at all for BIR. For physician practices, we relied on the judgment of respondents in a convenience sample of eleven practices. These judgments are subject to error, and a wider or more systematic sample of practices might have produced different estimates, although we observed consistency within our sample. Further, a sizable fraction of BIR activity in physician offices is accounted for by physician effort, and here we relied on estimates from a survey conducted by the AMA.

Our analysis applies to services paid for through insured arrangements. However, a substantial fraction of health care services in California are provided through self-insured arrangements. For self-insured services, we expect that hospital, physician, and insurer BIR would be similar to our estimates here, except that health plan profits might be smaller (most self-insured employers use health plans or other third-party administrators to administer health benefits, and these contractors need to make some profit to stay in business).

Our analysis excludes several sources of BIR administrative costs: external brokers’ fees, oversight provided by employee benefits staff, and administrative costs of independent practice associations (IPAs). It thus understates total BIR costs in the private insurance system. Also, our analysis of total BIR costs in privately insured hospital and physician care understates and mischaracterizes the distribution of systemwide BIR costs. Most importantly, hospitals and physicians incur BIR costs on all care, not just privately insured care. Thus, the total BIR burden on providers (in dollar terms) is well above that on private insurers—perhaps twice as much. Further, there are substantial, if less well understood, BIR costs in other care settings, in which one-third of all health care spending occurs.

Simplifying our methods of paying health care providers would reduce BIR, although anticipating the exact magnitude of savings is difficult. If a single-payer system of health care financing were adopted, a substantial portion of BIR could certainly be eliminated; less comprehensive financing reforms might yield smaller savings. We are just beginning a project studying a large multisite physician practice, to better understand what fraction of BIR costs at that practice could be eliminated with a variety of simplifications in financing rules or procedures.

Uncertainty about BIR spending and value. There is much uncertainty about the extent to which current BIR spending creates value. BIR activity could produce two types of value. First, it could lead to lower rates of payment per unit of service than would occur without it. This would result in a shift of resources from providers to patients, resulting in no net increase in social welfare, although a potentially significant redistribution.¹² Further, to the extent that rapidly increasing rates of health care spending create strains on public and private financing systems that lead to other negative results (such as increasing numbers of uninsured people or crowding out of other public-sector funding priorities), restraint in the growth rate of health care spending could do more than simply redistribute income from providers to patients. Partially counterbalancing the welfare gains from lower payment rates, a potential negative effect of slowing payment rates is reduced access to physicians.¹³ Second, BIR activities could result in a reduction in the use of health care services (through agreements reached between payers and providers on the services that will be paid for). To the degree that these reductions occur with no decline (or perhaps even an improvement) in the outcomes of care, the BIR services create value. However, if effective care is reduced, health care value is lost.

Net impacts of BIR. The net impacts of BIR are poorly understood. If BIR activities are, or are perceived as, simply a labor- and capital-intensive mechanism to move money around in a highly fragmented health care financing system, they are attractive targets for efficiency-increasing attempts at health care reform, because they consume resources that would be better spent on health care.

The empirical record is complicated. In comparison to the health care systems of other advanced economies, it is difficult to argue that the United States has an efficient health care system: Its high level of clinical health care spending does not seem to be matched by superior outcomes. However, competition among insurers from 1993 to 2000 almost certainly contributed to constraint of the rate of health spending growth in that period. Although BIR might have increased during the 1990s, the reduction in health care spending was greater than this increase in BIR and is likely attributable in part to the incremental BIR. Some argue that the slowing of spending growth led to negative effects on quality and a reduction in consumer welfare, although the evidence is complex. Further, very rapid spending growth returned after 2000. Regardless of the experience of the 1990s, prospects that insurer competition will foster substantial improvements in quality or efficiency in the coming years are uncertain at best. To the extent that competition does not foster cost control and quality improvement, and to the extent that a simpler system with fewer insurers would allow a reduction in BIR, then reductions in BIR are an attractive target for reform initiatives.

Editor's Notes

 
James Kahn (jgkahn{at}ucsf.edu) is a professor of health policy and epidemiology at the Institute for Health Policy Studies (IHPS), University of California, San Francisco (UCSF). Rick Kronick is a professor and chief of the Division of Health Care Sciences, Department of Family and Preventive Medicine, University of California, San Diego, in La Jolla. Mary Kreger is a senior research associate in IHPS, UCSF. David Gans is director, Practice Management Resources, at the Medical Group Management Association in Englewood, Colorado.

This work was funded by Grant no. 00-4071 from the California HealthCare Foundation. The authors thank the individuals at medical practices that responded to their survey, and also Milliman USA for providing data on insurer administrative costs.

NOTES

Top Study Data And Methods Study Results Discussion NOTES

 

1. See, for example, S. Woolhandler, T. Campbell, and D.U. Himmelstein, "Costs of Health Care Administration in the United States and Canada," New England Journal of Medicine 349, no. 8 (2003): 768–775[Abstract/Free Full Text]; S. Woolhandler and D.U. Himmelstein, "Costs of Care and Administration at For-Profit and Other Hospitals in the United States," New England Journal of Medicine 336, no. 11 (1997): 769–774[Abstract/Free Full Text]; and S. Woolhandler, D.U. Himmelstein, and J.P. Lewontin, "Administrative Costs in U.S. Hospitals," New England Journal of Medicine 329, no. 6 (1993): 400–403.[Abstract/Free Full Text]
2. P.M. Danzon, "Administrative Costs: Answering the Critics," Health Affairs 11, no. 2 (1992): 231–233[CrossRef][Medline]; H.J. Aaron, "The Costs of Health Care Administration in the United States and Canada—Questionable Answers to a Questionable Question," New England Journal of Medicine 349, no. 8 (2003): 801–803[Free Full Text]; and A.K. Gauthier et al., "Administrative Costs in the U.S. Health Care System: The Problem or the Solution?" Inquiry 29, no. 3 (1992): 308–320.[Web of Science][Medline]
3. Woolhandler et al., "Administrative Costs in U.S. Hospitals"; and J.C. Robinson, "Use and Abuse of the Medical Loss Ratio to Measure Health Plan Performance," Health Affairs 16, no. 4 (1997): 176–187.[Abstract]
4. K.E. Thorpe, "Inside the Black Box of Administrative Costs," Health Affairs 11, no. 2 (1992): 41–55[CrossRef][Medline]; D.J. Shulkin, A.L. Hillman, and W.M. Cooper, "Reasons for Increasing Administrative Costs in Hospitals," Annals of Internal Medicine 119, no. 1 (1993): 74–78[Abstract/Free Full Text]; Woolhandler et al., "Administrative Costs in U.S. Hospitals"; and Danzon, "Administrative Costs."
5. See, for example, Institute of Medicine, Crossing the Quality Chasm: A New Health System for the Twenty-first Century (Washington: National Academies Press, 2001).
6. A complete description of methods is available in a technical supplement, online at content.healthaffairs.org/cgi/content/full/24/6/1629/DC1.
7. Medical Group Management Association, Cost Survey: 2001 Report Based on 2000 Data (Englewood, Colo.: MGMA, 2001).
8. P. Zhang and S.L. Thran, eds. Physician Socioeconomic Statistics, 1999–2000 Edition (Chicago: Center for Health Policy Research, American Medical Association, 1999).
9. Ibid.
10. The source of our task breakdown was Lewin Group, Cost and Coverage Analysis of Nine Proposals to Expand Health Insurance Coverage in California: Final Report (Sacramento: California Health and Human Services Agency, 31 March 2002).
11. The specific formula and other technical details are available in the technical supplement; see Note 6.
12. M.V. Pauly, "Should We Be Worried about High Real Medical Spending Growth in the United States?" Health Affairs, 8 January 2003, content.healthaffairs.org/cgi/content/abstract/hlthaff.w3.15 (29 August 2005).
13. L. Backus et al., "Specialists" and Primary Care Physicians’ Participation in Medicaid Managed Care," Journal of General Internal Medicine 16, no. 12 (2001): 815–821.[CrossRef][Web of Science][Medline]

                      What's this?

This article has been cited by other articles:

				K. Davis, C. Schoen, K. Shea, and C. Haran Aiming High for the U.S. Health System: A Context for Health Reform. J. Law Med. Ethics, December 1, 2008; 36(4): 629 - 643. [PDF]

				L. L. Berry, A. M. Mirabito, S. Williams, and F. Davidoff A Physicians' Agenda for Partnering With Employers and Insurers: Fresh Ideas Mayo Clin. Proc., December 1, 2006; 81(12): 1592 - 1602. [Abstract] [Full Text] [PDF]