“If health care is going to spend less, some inputs will need to be paid less. This post explores the possibility of saving money by reducing the administrative costs of health care.
Reducing administrative costs is attractive for several reasons. Administrative costs are high, perhaps a quarter of health spending, so reductions in administrative costs could yield a good deal of savings. Further, the goal of medical care is clinical care, so reducing administrative staff likely has a smaller effect on quantity and quality of care than would reductions in clinical staff. Finally, excess administrative hassles adversely affect peoples’ ability to receive care, so reducing administrative hassles could improve the timeliness of care received.
[..] Not all administrative costs are wasteful. For example, money spent on electronic medical record systems can improve the quality of care. Thus, one needs to tread carefully on the administrative end, as with everything else in health care.
[..] Some of the failure of health care to standardize is a result of the fact that health care is more complex than retailing. A box of breakfast cereal rarely changes. In contrast, medical procedures change frequently and in subtle ways. But that is not the whole story.
[..] In the 1970s, there was a growing realization that the cost of sending money from bank to bank, which was then done by paper, was too high to meet the demand for transfers. The Federal Reserve imposed order by establishing a uniform set of criteria for transferring money electronically. All financial institutions that wish to send or receive funds must do so the same way. The net effect is that over $50 trillion is transferred across financial institutions annually, at minimal administrative expense.
Health care is more complex than banking, but the principal remains: if we want standardization, the federal government is going to have to impose it.
The reality is that in health care, there is little incentive to standardize, and often incentives to avoid standardization. There may be three or four different insurance companies in the market, none of whom wants to change their internal processes to match any others. Changing processes is expensive, and when one insurer does it without others doing the same, they will be at a competitive disadvantage. Further, they do not have the right to see what other insurers are doing, nor do they have any influence in decisions made by the other insurers.
Health Care Automated Clearing House
[..] In banking, the transfer system is built off automated clearinghouses (ACH). One could do the same in health care; call it the Health Care Automated Clearing House (HC-ACH). There are two principal actors in a clearinghouse. There is an organization that sends information back and forth—money in banking, claims and clinical information in health care. Such an organization needs to have key security processes in place. In addition, there is an organization that sets standards for exactly how things must be coded. In banking, this is done by Nacha (formerly the national association of clearinghouses). In health care, there is already an equivalent organization: the Council for Affordable Quality Healthcare (CAQH). Through a consensus process, CAQH sets standards for transfer of information between payers and providers, some of which has already led to significant savings.
As in banking, the cost of the HC-ACH would be paid by small fees assessed on member insurers. These fees would give insurers a stake in making sure the system operates efficiently. Further, a dedicated funding stream would relieve the organization’s ability to operate from the whims of federal funding. The cost of intermediary organizations is not large. I estimate that it could operate on a budget of about $300 million annually, roughly 8 cents out of every thousand dollars spent on medical care. Public and private insurers and provider systems would have to pay one-time costs to configure their computer systems to meet the requirements, although they already employ large numbers of people to maintain different billing processes. Within a short period of time, the savings should vastly outweigh the costs.
Streamlined Information Flows
Many other administrative costs in health care are a result of the fact that information does not flow seamlessly among providers and between providers and payers. Consider two examples.
First, quality metrics are often not based on information in the EMR or require information from different medical records which are hard to combine (for example, a hospital system and a physician system). If all information were easily transferrable and quality metrics were based on that accessible information, the burden of quality reporting would be significantly reduced.
Second, prior authorization is needlessly complicated by the fact that large amounts of information need to be transferred from providers to insurers. For example, an insurer may want proof that a patient meets certain clinical criteria before a radiology test is approved. To document this, the provider’s office must fax information from the medical record to the insurer, who then reviews it all before reaching a decision. A more streamlined way to do this would be to have the provider’s EMR system attest that the patient meets the relevant criteria, avoiding all the human interaction.
Streamlining information flows will require government involvement for the reasons noted above; private organizations will oppose many types of information flows. The federal government has recognized this and started to act—too late, but better late than never. Rules have been enacted or contemplated for information at various points in the system. More such rules will be needed, and policing will be needed to make sure that unnecessary roadblocks are not constructed.
Attaching a Price to Administrative Complexity
Rules and policing are important, but economic incentives can play a large role alongside them. Consider the example of prior authorization. In the current environment, insurers are free to impose any prior authorization requirements they wish, subject only to the decision of providers not to contract with them. That threat is relatively weak when directed at a large insurer; no provider can do without a significant share of patients in the area.
What is missing from this market is a price. By imposing prior authorization rules, the insurer is causing the provider to spend money on compliance. But the insurer does not pay for that cost. A natural strategy is to make the insurer pay the cost. There are estimates of the costs of prior authorization to providers—nearly $11 per case when done manually, according to CAQH—that could be used. In this way, insurers would have a financial incentive to consider whether the amount they save from imposing prior authorization is greater than the cost to the system of conducting the prior authorization.
The same rule could work with quality measurement. Policymakers could design a standard set of quality metrics, ideally based on electronically stored information. Insurers would be allowed to require other measures, but they would have to pay providers for the costs imposed by collecting the additional information. Thus, prices would help to temper unnecessary variation.
What Could Be Saved?
Single-payer systems involve the least administrative expense, perhaps two-thirds below the US level. Many single-payer systems involve very little documentation. Of course, single payer involves many other tradeoffs that may not be appealing.
Among multi-payer countries, administrative costs are perhaps half the US level. One benchmark is thus for the US to reduce administrative costs in half. If we did so, total medical spending could be reduced by about 10 percent. Even if only half that was achieved, the savings would still be immense.
[This post summarizes David’s paper for Brookings, published in May 2020. In that paper, David suggested annual savings in the $50-75 billion range. For comparison, Shrank et al.‘s 2019 JAMA article estimated these annual savings by targeting the following sources of waste:
- Failure of care delivery – $44.4-97.3 billion
- Pricing failure – $81.4-91.2 billion
- Failure of care coordination – $29.6-38.2 billion
- Fraud and abuse – $22.8-30.8 billion
- Overtreatment or low-value care – $12.8-28.6 billion
Full article, Cutler DM. Health Affairs Blog, 2020.10.2