'It is almost impossible to buy a traditional indemnity health plan in New York. Most of the insurance companies don't offer them any more and those that do are unaffordable for all but Bill Gates'

The university where I work had its health insurance premiums raised by almost 40 per cent this year because of 'bad claims experience' - in other words, people used the insurance they had. I was appointed to a committee to look into how the costs could be lowered and what could be done to prevent future increases. One of the few benefits of being a professor of health policy is the ability to think about the 'big picture' rather than the actual nuts and bolts of healthcare issues, but this has allowed me to look at it from the other side.

I learned that it is almost impossible to buy a traditional indemnity health plan in New York. Most of the insurance companies don't offer them any more and those that do are unaffordable for all but Bill Gates.

Indemnity plans are those where you get services from the doctor or facility of your choice, pay the bill yourself and are then reimbursed by the insurance company. This is the kind of plan we had. In theory, and in reality as well, there is almost no management in an indemnity policy except for the need for prior approval of elective surgery and steep co-payments and deductibles.

I also learned that the health plans allow for infinite customisation of benefits and limitations. For example, our plan could be underwritten prospectively or retrospectively, the level of deductibles, co-insurance and co-payments could vary, as could the services covered. We could also add different dental, pharmaceutical and eyesight services. This variety, which seems wonderful on the surface - and allows employers to tailor plans to their needs and their ability to afford them - creates havoc when it's time to pay the claims and also increases administrative costs.

There is a clear difference between the sales force which markets the policies and the claims payers who have to pay the resulting bills. The sales people try to sell their policies based on their comprehensiveness and ease of use. The claims payment experience often turns out to be quite different. Because there are so many variations in plans, each claims processor must know the intricacies of each one. When they delay or deny a payment, people are not correctly reimbursed or the doctors, hospitals and laboratories don't get paid.

One of the reasons we looked for a new insurance carrier was that the service provided by our current one was so awful. The reality belies the hyperbole of how wonderful the information systems are. What most surprised me was the reluctance of the sales people to sell managed care. I was expecting, perhaps naively, that the companies would tell us that the reason our premiums were increasing was because we didn't have managed care, and that in return for serious controls on the use of healthcare our costs would come down. The four firms we interviewed would not sell an indemnity policy, but they would not say that what they were selling was managed care either.

They would talk about how large their network of doctors was and how they had contracts with almost every hospital in the metropolitan area. They would say that they almost never deny care for any reason, which seems to fly in the face of what most people believe occurs in managed care. The only difference in the policies would be whether you self-referred within their network, went completely outside their network or worked through a primary care doctor in a smaller network.

For managed care to be successful, people have to understand that the only way to keep medical costs - and therefore premiums - down is by restricting care, reducing fees or providing care in a more efficient manner. Although no one in my university is enamoured with the idea of paying higher health insurance premiums (and individual staff pay no more than 9 per cent of the total cost of the premium), people are even less excited about having to change doctors or having their ability to seek care restricted in any way. Moreover, the increased premiums mean that there will be less money available to the university to pay for salary increases, financial aid for students or new computers and books for the library.

Ultimately, we settled for an insurance carrier that will only raise our premium 25 per cent above the current level and will substitute a preferred provider option for our indemnity plan, while continuing our open-ended health maintenance organisation and our ordinary HMO. In a PPO, you can go to any hospital or doctor, but if you chose one in the network of the plan, you pay only a small co-payment rather than 20 per cent of the total cost.

The open-ended HMO (also called a POS or point-of-service plan) gives you a primary care doctor and the ability to have all your care - for a small co-payment - in the network, but you can also go outside the network if you want something else. The regular HMO restricts you to the primary care doctor and a limited network.

The current congressional debate over the need to regulate managed care and to prevent its excesses has certainly affected my colleagues, who are nervous about getting into a health plan that will restrict them in getting care they believe they need. While the issues in Washington are the ability to sue for denied services and the ability to get compensation for pain and suffering, the issues at my university are whether we can go to the doctor we want or get care from a particular specialist at a particular hospital.

By not dealing more directly with this issue, the managed care companies are looking for trouble. Large numbers of dissatisfied customers will complain to their employers, who will be forced to find new companies to deliver the care. The real debate about the trade-off between freedom of choice of doctor or hospital and cost of health services has yet to be played out. Until it is, we will neither reduce healthcare costs nor create a more effective and efficient healthcare system.