How to Fix Health Care

By Dan Crippen
Sunday, May 1, 2005

In just 25 years, federal spending for Social Security, Medicare and Medicaid (which was originally designed for the poor but increasingly also supports long-term care for the elderly) will about equal what we now spend on the entire federal government. Paying for those programs, without reform, will take tax increases of nearly 10 percent of gross domestic product -- the equivalent of payroll taxes of 30 to 35 percent.

Social Security today is the chief topic of conversation, but it's no secret that health care costs for the elderly will increase much more than retirement payments and much more quickly.

If we are to provide health care for seniors in superior and ultimately cheaper ways, we must face critical facts that point the way to effective, long-term Medicare reform:

First, the lion's share of Medicare spending is going for a relatively small number of people. Second, we are wasting time and money by not having a coordinated care system for these big users. Third, we lack the information needed to guide our caregiving. Fourth, we continue to drive up costs by overusing hospitals and preventing nurses and technicians from doing routine work that doctors now needlessly perform.

Let's start with the cost breakdown. Just 10 million of the 40 million Medicare beneficiaries are burning through 90 percent of the program's costs every year. That's where big changes are needed to hold down costs and improve care. But to build political support for reform we also must do something for the 30 million who are responsible for just 10 percent of the costs.

These relatively healthy people don't cost enough for the government to be following them around with a rulebook or forcing their physicians and pharmacists to file mountains of paperwork. Instead, for the majority of them, we should introduce "smart cards" to serve as virtual medical coordinators. By using the cards, recipients could go to any doctor, get any lab test and fill any prescription -- with the appropriate co-pays -- and the transactions would be electronically recorded (and could be monitored to identify patients who are headed for trouble). They wouldn't be constrained by a government-defined "benefit package" or separate pharmaceutical benefit, and they would enjoy all the freedom of medical savings accounts.

But unfortunately this wouldn't save much money and possibly could cost a little more. So what to do about the 10 million patients who are driving most of the costs? Some are in their last weeks or months of life. But most of the program's costs are for patients who are chronically ill and in need of extensive care. They typically have multiple chronic conditions such as diabetes, heart disease, high blood pressure and lung disease. They routinely consult a dozen or more doctors and fill 50 prescriptions per year. They are often hospitalized at least once a year. And no one is in charge of coordinating their care, which results in overuse of medical services and conflicts in treatments and drugs.

We need to know more about these patients. Only recently have government agencies begun to assemble the mountains of data the government collects into patient records. When my father visits a physician, it is treated as a single, separate transaction in the record. Any subsequent tests, treatments or visits are similarly reported, each with its separate record. Rarely has anyone examined what he was being treated for, what the results were of those treatments or the costs he incurred. And we know little about treatments not covered by Medicare, such as pharmaceuticals. As we collect data, we can begin to assess the state of knowledge on treatment of multiple conditions to determine the best forms of treatment, including pharmaceuticals, and the best settings for treatments.

We should also begin to develop and test compensation systems that encourage physicians to change their practice patterns. Much of the past three decades has been spent trying to modify patient behavior by changing deductibles and co-pays -- with only marginal effects on health costs. The truth is that most of the high costs incurred in Medicare are for procedures that patients do not consider discretionary -- hospitalization and inpatient procedures and testing. When a doctor tells my father he needs to be hospitalized for a series of cardiac tests, my father doesn't think about his deductible and co-pays -- he goes to the hospital.

Doctors should have incentives to assume the role of care coordinators for the chronic users of Medicare in an effort to improve care and hold down costs. We should start reducing the utilization of hospitals to curb costs without lowering the level of care provided. The current revolving door between nursing homes and hospitals is unnecessary, expensive and a burden to patients. Some basic tests and care procedures, such as diagnosis and treatment for dehydration, can easily be performed by other health care workers in nursing homes and other settings, if doctors will allow it. Hospitals need not worry about emptying out. The number of elderly is about to double; we could cut admission rates in half and still keep utilization at today's levels.

With proper incentives, we can have physicians coordinating the entire care regimes of the chronically ill, regulating the prescriptions and doing everything in their power to provide the best treatment without using the hospital as a default. The patient will get better care, and Medicare will recoup big savings.

President Bush is to be commended for opening the debate on Social Security, but we can no longer ignore the elephant in the room that is Medicare. It's too important for everybody.

The writer is a former director of the Congressional Budget Office and was an economic adviser in the Reagan administration.

© 2005 The Washington Post Company