General practice is the specialty that learned less medicine can mean better care.

In July 1979, the British Medical Journal published a trial about helping people stop smoking. Smokers who wanted help to quit could go to a special clinic. But the whole UK had only about 50 of these clinics. The trial asked whether general practitioners could help instead, simply by advising their patients to stop. It included all 2,138 smokers who saw 28 general practitioners in London over four weeks. Two groups got no advice. One group heard their GP advise them to stop smoking. The last group also got a leaflet and a warning that the GP would check on them. The researchers counted the smokers who stopped in the first month and were still not smoking a year later. They found 0.3% of the smokers who got no advice. They found 5.1% of the smokers who got advice, a leaflet and the warning. The authors worked out that one GP could expect about 25 lasting ex-smokers a year. Every GP in the UK together could produce more than half a million a year. A few words from a family doctor could do what the clinics could not.

Almost eighteen years later, in March 1997, the same journal published a trial about sore throats. Doctors often prescribed antibiotics for a sore throat, but nobody knew how much the antibiotics helped. The trial took 716 patients with a sore throat at 11 general practices. One group got antibiotics for ten days. A second group got no prescription. A third group got a prescription to use only if the throat was not settling after three days. The third group left 69% of those prescriptions unused. All three groups recovered at the same pace. The antibiotic group had one fewer day of fever. The antibiotics mostly changed what patients believed. Belief that antibiotics work reached 87% in the antibiotic group. It reached 55% in the group given no prescription. The antibiotic group also planned more visits to the doctor for their next sore throat. Holding back the antibiotics had cost patients one day of fever, and it made them less likely to come back for the next one.

Almost twenty-one years later, in February 2018, The Lancet published a trial about type 2 diabetes, run inside ordinary general practices. Type 2 diabetes was seen as a disease that needs treatment for life. The trial asked whether weight loss, managed by the practice, could put it into remission. It took 306 people from 49 practices in Scotland and England. All of them had been diagnosed within the last six years. The practices in the programme stopped their patients’ diabetes and blood pressure pills. The patients lived on a formula diet of about 850 calories a day for three to five months. Then they slowly went back to normal food, with support to keep the weight off. The other practices gave the best care the guidelines allowed. The researchers checked for remission after a year. Diabetes went into remission in 46% of the people in the programme. It went into remission in only 4% of the people given usual care. Remission followed the weight loss. It reached 86% of the people who lost 15 kilograms or more. Serious side effects affected 4% of the people in the programme. They affected 1% of the others. A diet run by the practice had put a lifelong disease into remission.

These are three of the thousands of studies that shaped modern general practice. They taught general practice that a few words of advice could make smokers quit, that holding back antibiotics for a sore throat cost patients very little, and that a diet run by the practice could put diabetes into remission, and each time the answer was in a paper years before it was in a textbook.

It is still happening. Recently, in June 2025, Nature Medicine published a trial about blood pressure and dementia. Dementia is a leading cause of death and disability worldwide. The trial asked whether lowering blood pressure could prevent it. It took 33,995 people aged 40 or older with uncontrolled high blood pressure, in 326 villages in rural China. Trained community health workers in 163 of the villages started and adjusted blood pressure pills. Primary care doctors supervised them. The health workers aimed for a blood pressure below 130 over 80. The other 163 villages kept their usual care. The health workers lowered systolic blood pressure by 22.0 mm Hg more than usual care did, over four years. Dementia was 15% less common in their villages. Serious side effects were less common too. Treating blood pressure had also prevented dementia. Practice follows the literature as it goes. It is a good culture. It has exactly one cost.