DEAR READERS: In yesterday's column, I responded to a reader's question about acute pharyngitis -- inflammation of the throat caused by infection with bacteria or viruses.

I was taught that diagnosing and treating a patient with a sore throat was not complicated: The sore throat was caused either by Group A streptococcus ("strep," a kind of bacteria) or by a virus. If a throat culture showed strep, you treated it with penicillin. Simple. But in my view (some colleagues disagree), it's not that simple.

The risk from an untreated infection with Group A strep is much lower today in the United States than it was 70 years ago. That means that the value of treatment is reduced. But the chance of side effects from the treatment - antibiotics - is not reduced. Penicillin can cause side effects ranging from a bad rash to a life-threatening drop in blood pressure. A doctor never wants to prescribe a treatment when its risk is worse than the risk from the disease.

If a doctor knows for sure that a patient has a strep throat, the value of antibiotic treatment today is still greater than the risk. The problem is that antibiotics are of no proven value if a patient does not have a strep throat. In addition, it is hard for a doctor to know for sure that a patient has a strep throat without a positive throat culture. An eminent pediatrician, Dr. Burtis Breese, once gave his colleagues some wise advice: "If you are entirely comfortable selecting which patients with pharyngitis to treat with penicillin, perhaps you don't fully understand the situation."

As I mentioned in yesterday's column, the results of a throat culture take one to two days. It's sometimes hard to reach the patient when the results return. For that reason, doctors had long yearned for a test that would reveal whether the patient had a strep throat while he or she was still in the doctor's office.

About 30 years ago, rapid strep tests were developed. The latest versions of the test are fast: Results come back within 15 minutes. And the tests appear to be highly accurate in detecting Group A strep. So this seemed like a major advance.

But, again, it's not that simple. First, the rapid tests are quite accurate, but not perfect. Second, we've learned that other kinds of strep besides Group A strep, and other kinds of bacteria entirely, can cause acute pharyngitis. The rapid strep tests don't detect any of these bacteria. Admittedly, we don't yet have solid evidence of the value of treating these bacteria.

Third, sometimes Group A strep live in our throat without causing any disease: There is no value in killing those strep. Finally, even when we test for all known bacteria, viruses and other microbes, we still can't find a cause of sore throat in about 30 percent of patients.

So there's still a lot to learn about a medical problem that causes more days lost from work than all labor disputes combined. Yet we've cut federal funding for medical research by 20 percent over the past decade.

Dr. Komaroff is a physician and professor at Harvard Medical School. To send questions, go to AskDoctorK.com, or write: Ask Doctor K, 10 Shattuck St., Second Floor, Boston, MA 02115.