Face beet-red, fists balled up, and back arching, the 6-week-old baby lay on the exam table screaming.

Had I not had my own baby with colic, I would hardly have believed a baby could cry that hard for hours on end, setting everyone's nerves on edge. The baby's mother came to me for advice on whether to medicate her baby.

She had searched the Internet and talked to several neighbors who all recommended Zantac, an acid blocker normally prescribed for adults with heartburn. The data on Zantac as a treatment for colic were mixed, and my own experience with my patients was not convincing. But I could hardly blame the mother for wanting some relief - any relief - from this distressing situation.

Children now take drugs for attention deficit hyperactivity disorder, depression, bipolar disorder, anxiety, vitamin D deficiency, iron deficiency, lack of fluoride, calcium deficiency, heartburn, slow gastric emptying, urinary backup, menstrual regulation, recurrent ear infections, constipation, bed wetting, headache, obsessive compulsive disorder, and anorexia - the list goes on.

Obviously, some of these drugs are essential, but physicians are increasingly putting young children on multiple medications that often are taken for weeks or months - sometimes years - and that's cause for concern.

Recently the American Academy of Pediatrics recommended diet modifications, exercise, and possibly medication for children over 8 years old with elevated levels of overall cholesterol, particularly the bad kind, known as LDL. High cholesterol is clearly a danger sign in an adult, and may be for a child. But for young people, the long-term consequences of elevated cholesterol - or cholesterol-lowering medications - are unclear.

Just because we have a drug to lower cholesterol or improve behavior or reduce anxiety, and just because adults are taking those drugs, that doesn't mean we should be giving them to children. When is a disorder part of the human condition and when is it something we really have to treat? I think we've lost sight of that line.

Yes, some people will blame the problem on pediatricians like myself. And yes, we are part of the problem. For instance, we used to put children with recurrent ear infections on low-dose antibiotics for months at a time to prevent infection; we now know that they are unnecessary and may increase a child's risk of later antibiotic resistance.

But there is plenty of blame to go around, including the specialists who feel the need to "do something," the parents who are searching for solutions, and, of course, the drug companies that remind us in every media outlet - and in my own office with donations of pens and free lunches - that there is a "cure" for everything.

One recent example is the push to treat girls 13-16, whose periods come at unpredictable times, with Yasmin, a pill made by Bayer Schering Pharma. Yes, we can give them the pill to regulate their periods and reduce cramping, acne, and mood swings. But should we treat something that's so clearly part of normal physiology?

Given the ever-changing science of medicine, it is prudent to adopt new medications for children carefully, especially when they are to be given for a long period of time. All medications have potential side effects. I often see children with diarrhea, rashes, fatigue, abdominal pain, weight loss or weight gain, headache, vomiting, or mood disorders - that turn out to be side effects from medications. And, of course, long-term data on these medications are often unavailable for children.

I have also seen children with attention deficit hyperactivity disorder go from a wall-bouncing, disruptive whirling dervishes to a thoughtful, calm students with improved self-image, thanks to medication. And in my early days of practice, it was not uncommon to see children die from acute lymphoblastic leukemia; today, if diagnosed early, most of these children do well. "Big pharma" has given us wonder drugs that do help children.

But since heart disease, for instance, is still a puzzle - does cholesterol cause cardiac illness or is it underlying inflammation - we had best go slowly before sentencing children to a lifetime of medication. As the medication list for our young patients grows longer, we need to continually analyze the data and consider the long term consequences of each new recommendation.

Drugs are lifesaving, pain-saving tools, but sometimes they're not necessary, and where we draw the line is critical.

Dr. Victoria Rogers McEvoy is chief of pediatrics of the Mass. General West Medical Group and assistant professor at Harvard Medical School.