When he was 9 years old, our middle son began to behave oddly. Normally talkative and happy, he seemed to have no energy and complained often of feeling sick. We had him examined by the pediatrician but found nothing. We took him to a therapist, who thought he seemed depressed.

A few weeks after the first meeting with the therapist, David began to show symptoms we understood. He started to get extremely thirsty and had to run to the bathroom frequently. This time, the pediatrician had no trouble with the diagnosis: Type I diabetes. We went straight to the emergency room and into a new life.

There are worse things, we reminded ourselves from the start. A trip to Children's Hospital is pretty much guaranteed to make you count your blessings. That much having been said, diabetes is no picnic. There are up to 3 million Americans with Type I -- an auto-immune disease in which the body destroys its own insulin-producing cells in the pancreas -- and as many as 20 million with Type II, a disorder in which cells lose the ability to absorb insulin.

Type I diabetics have to think about their condition every day, all day. David wears an insulin pump, which replaces injections. He must test his blood glucose (BG) with finger sticks a minimum of five times a day and count the carbohydrates he eats so that he can calculate the proper dose of insulin he needs.

He must also (and this is a pitfall for kids) remember to give himself insulin for every morsel he eats, and change the infusion site of the pump (a catheter that is inserted under the skin with a needle) once every three days. He must be aware that physical activity, illness, stress and temperature can affect blood glucose levels. If BG goes too high, his risk of long-term complications like blindness, nerve damage, amputations, kidney failure and heart disease rises. If his blood sugar goes too low, either because he doesn't eat or because he gives himself too much insulin, he can lose consciousness and even die.

When I mention to people that David lives with diabetes, I often hear cheery predictions that medical research is on the cusp of finding a cure. Doctors, particularly those in other specialties, seem quite confident of this. But as I've learned, medical research is not some self-propelled machine that rolls ever forward. Drug companies invest in research, as do governments, universities and private foundations. But companies often require assurance that a new technology will be covered by insurance before they invest too much. And finding a cure for Type I diabetes, which affects only a few million people, will not rank as high as finding a cure for more common ailments.

There has been a great deal of buzz about stem cell research and diabetes. The theory there is that if we could grow a new pancreas using our own stem cells, we could transplant a new organ without the need for damaging anti-rejection drugs. Maybe. But this prospect is distant at best.

On the other hand, we are tantalizingly close to a technology -- the artificial pancreas -- that, while not a cure, would dramatically improve the lives of diabetics (including many with Type II). Continuous glucose monitors (CGMs) have just begun to be marketed. Using a subcutaneous probe, they measure BG every five minutes without the need for finger sticks. So we now have two technologies: the CGM and the insulin pump, one that measures blood glucose and one that doses insulin. They are the two halves of an artificial pancreas. The brass ring will be getting the two to speak to each other.

The Juvenile Diabetes Research Foundation (www.jdrf.org) is about to begin a clinical trial to measure whether CGMs improve the overall blood sugar control of diabetics. Early indications are that they do. (Those who wear the devices are amazed at how little information they had been receiving from finger sticks alone.) Next the challenge will be to develop the algorithms to permit the CGM to speak to the pump -- and to do so safely. If the algorithm is not foolproof, the wrong dose of insulin could kill the patient.

David is now 13. He owes his life to Dr. Frederick Banting of Toronto, who discovered insulin in 1921. But his long-term prospects are in the hands of thousands of researchers, government officials, doctors and insurance company executives who will decide whether the next great leap in diabetes treatment is vital or not.