Often deciding when an infection might be viral or bacterial boils down to statistics. Someone with strep throat may have most of the positive signs – sore neck lymph nodes, fever, pus on the tonsil, no cough and be in the target age range… and still have a viral infection. The physician often treats with an antibiotic, even though the chance it will help is close to 50-50. I would tell this to the patient, who may opt for treatment or to wait two days for the throat swab. I do not consider this situation a placebo, there is a reasonable chance the antibiotic will help. If they have a sore throat and none of the above signs, the chance the antibiotic might help goes down to, say, 10% and I would not prexribe it without a positive throat swab. This is a differnet pre-test probability.
If an antibiotic is prescribed, there is a considerable difference between given penicillin, which treats most strep throats, versus strong antibiotics like Rocephin or Avelox whose use I reserve exclusively for very severe infections. Such antibiotics are used by some primary care providers to treat uncomplicted sinus and throat infections and this use is nearly always inappropriate.