If a patient has a functioning immune system, it generally isn't necessary for antibiotics to clear the whole infection, especially if they're not in the hospit

If a patient has a functioning immune system, it generally isn't necessary for antibiotics to clear the whole infection, especially if they're not in the hospital. The course will cull the infection long enough for the patient's body to acquire a sufficient degree of adaptive immunity to heal themselves completely. In some cases (tuberculosis), a typical patient's immune system can't be counted on to win, so cocktails are the norm. Bacterial resistance to antibiotics is not new; as Wiki notes, the genes are "ancient", and simply become more prevalent when antibiotics are used. http://en.wikipedia.org/wiki/Tuberculosis_management#The_standard_regimen http://en.wikipedia.org/wiki/Tuberculosis_management#The_sta... "The frequency of spontaneous mutations that confer resistance to an individual drug are well known: 1 in 10^7 for EMB, 1 in 10^8 for STM and INH, and 1 in 10^10 for RMP." So resistance is basically everywhere -- it becomes deadly when it is concentrated.