In the United States, the agency responsible for approving medications is the Food and Drug Administration. In its own words, it protects the public by assuring that "human and veterinary drugs, and vaccines and other biological products and medical devices intended for human use are safe and effective."
This doesn't mean the FDA itself conducts tests on drugs. Rather, drug manufacturers have to perform rigorous testing on any new medication, testing that checks for safety, side effects, how well the drug works, and whether it has more benefits than existing medications.
Generic drug manufacturers, on the other hand, only have to prove that their versions of brand name drugs are as safe and effective as the brand names.
The chronically underfunded FDA is always under a lot of pressure. People want drugs approved more quickly, yet they also want more scrutiny of drugs' safety - which can take much longer. Generally, the final phase of testing new medications lasts 6-8 weeks - far too little time to get any long-term data.
The debate, which has resulted in multiple changes to FDA procedures since Dr. Frances Kelsey refused to approve the infamous drug thalidomide for use in the US, still rages. (In other countries, thalidomide caused massive numbers of children to be born with severe birth defects. Dr. Kelsey was awarded the Distinguished Federal Civil Service Award by President Kennedy.)
A recent problem with a generic drug highlighted a problem both in the FDA's procedures - which will be changed - and in its need for more funds. The antidepressant Budeprion XL 300, a generic form of Wellbutrin XL 300, was found not to deliver equivalent results as the brand name. The generic had been approved on the basis that another form of Budeprion (made by Teva Pharmaceuticals) had passed all tests.
Reports that the drug was not working as well as the brand name began pouring in in January 2007, shortly after the generic's approval. These complaints were monitored for most of the year, and after 10 months the FDA ordered Teva to perform new testing.
Here's where the problems really began. Teva's testing was agonizingly slow due, they said, to problems enrolling enough people for the study. The FDA decided to sponsor its own testing in 2010, but first had to obtain funding, along with getting approval for it, designing the study, and recruiting volunteers before it could even begin. Results were not available for two years. Had funds been readily available, perhaps this time frame could have been reduced.
As a result, it was more than five years from when the first complaints appeared until Budeprion XL 300 was withdrawn from the market on October 3rd, 2012.
The problems of public expectations, changes in legislation, and lack of funding for the FDA aren't going to go away anytime soon. However, the story of Budeprion XL 300 does show that the voice of the public regarding specific drugs is heard. You yourself can report a problem with a medication here: FDA Adverse Reporting System.