The District of Columbia government spends extravagantly on health care, but, unfortunately, what it gets in return isn't worth it.

"On a per capita and per patient basis, across almost any parameter, the District has been spending more for health care than any other jurisdiction in the country, including cities with similar demographics," a 1995 Blue Ribbon Panel on Health Care Reform Implementation reported. That's still true. The District spends $1.3 billion a year in this area--about $2,500 for every city resident, compared with a national per capita average of about $1,000. Meanwhile, our infant mortality rate is the highest of any large American city, our AIDS rate is the third-highest, and our cancer rate is the seventh-highest.

Mayor Anthony A. Williams has been trying to figure out how to improve this dismal picture. Three months ago, he set up a commission to examine the problem. Having held several public hearings, the commission is expected to issue recommendations within weeks that can be considered by the D.C. Council as part of the budget process. So this is a good time to step back and consider what changes are needed.

The most important step the District could take would be to extend health insurance coverage to the estimated 75,000 District residents who don't have it. This may seem obvious, but it isn't. After all, we have a large public hospital (D.C. General) and a variety of clinics, both public and private, that people can go to without paying anything, so why bother with insurance?

The answer is that insured people are more likely to get medical help early on, when problems can be dealt with before they become serious. A 1995 study conducted by the District Hospital Association found that the hospital admissions of uninsured people were three times as likely to be "avoidable" as those of insured people. So insurance is important.

And it won't require a lot of new money. We could finance universal coverage largely by reallocating a portion of current expenditures. Much of what we're spending now takes the form of lump-sum payments to health care providers that are supposed to serve the uninsured. The problem is that most of these providers are hospitals, and the most expensive way to see a doctor is from a hospital bed.

The District has one of the most hospital-based health care systems in the nation. Consider the Medicaid program, which I ran for four years, and which insures more than one-fourth of the District's population. In 1996, the latest year for which detailed figures are available, Medicaid recipients across the United States used 560 days of inpatient hospital care per 1,000 enrollees. The number for the District was 1,360. In other words, District Medicaid recipients spent almost two-and-a-half times as many days in the hospital as the national average. Meanwhile, of every Medicaid dollar, the District spent less than half as much as the rest of the nation on physician services, home health care and prescription drugs, and 50 percent more on hospital care. In effect, we're shortchanging those providers who can prevent people from getting sick so we can pay our oversized hospital bill.

How do we begin to fix this? The District pays D.C. General and the district's public clinics about $100 million a year to serve the uninsured, in return for which it receives services worth about $50 million, according to a report prepared by staff of the District's financial control board. In addition, private hospitals such as Children's and Howard University Hospital receive more than $40 million a year to serve the uninsured. All of this money could be used to buy health insurance for those who do not have it. Once they were insured, their hospital visits would still be covered, as well as visits to doctors and other primary care providers. Because these providers would see more paying customers, we could expect to see more of them opening up offices in the poorer parts of town to accommodate the new demand.

There's no question that cutting their subsidies would be hard on the District's hospitals, which are already in financial trouble. Business is down because many people have moved out of the city, suburban hospitals have grown, and managed care plans seek to limit their members' hospital stays. As a result, D.C. hospitals now operate well under capacity, and many of them are losing money. The problem is certainly not unique to Washington, but while other cities have tried to address it through hospital closures, mergers or downsizing, our hospitals have resisted such steps, as documented in a recent report from the Urban Institute.

A year ago, Williams proposed taking some of the hospital subsidy money and using it to expand health insurance for D.C. residents, but the council voted it down. That was not surprising: Hospital leaders are politically sophisticated; important business and civic leaders sit on hospital boards; and the council is made up of politicians who need the support of such people. A similar proposal would probably meet the same fate today, because the people who stand to benefit from reform are not well-organized, and lack the connections of the hospital leaders. The challenge is to work out a compromise that can get political support and still move us in the direction of broader coverage.

Fortunately, the District is about to start receiving $48 million a year as a result of the settlement of class-action lawsuits against the tobacco industry. Every interest group in town is lining up to stake its claim to this money, but surely it makes sense to use at least some of it to improve the public health (as opposed to refurbishing school buildings, for instance, as the mayor suggested last year).

So here's my proposal: The District would buy health insurance for the uninsured (up to an income-based limit--there's no need to subsidize rich people who haven't bothered to buy insurance) at an annual cost of between $80 million and $120 million. That money would come more or less equally from three sources: the tobacco dollars, the subsidies now going to private hospitals, and the subsidies going to D.C. General and the public clinics. While the hospitals would lose most of their special funding, they would benefit from the fact that just about everyone coming to them would have health insurance. That's a fair trade. Over time, as more resources move into preventive care, hospital revenues would decline, and there would likely be some reduction in hospital capacity as a result of closures or consolidation. But this makes sense, since the District has more hospital capacity than it needs.

The goal of the mayor's commission should be to work out an acceptable plan to finance expanded insurance coverage, and do it quickly. A few years down the road, the economy may sour, and we won't be in a position to act. Let's put aside our differences, and grab this historic opportunity to both improve health care and get a better bang for the District's dollar.

Paul Offner, director of the District's Medicaid program from August 1995 to October 1999, is a professor at Georgetown University's Institute for Health Care Research and Policy.