"We had hoped to see improvements in behavior and function in activities of daily living, but these did not occur either".
"Whilst previous smaller studies have suggested that exercise can prevent or improve cognitive decline in people with Alzheimer's disease, this robust and very large study provides the most definitive answer we have on the role of exercise in mild-moderate Alzheimer's disease", he told via the Science Media Centre.
There are about 10 million new dementia cases each year.
Although a short-term improvement in physical fitness was reported among patients in the exercise group, cognitive impairment after 12 months had declined a similar amount in both groups. They were recruited through memory clinics - specialist services that help people who have problems with their memory - and GP surgeries.
The ones chosen for exercise took part in twice-weekly, 90-minute gym sessions for four months, along with a one-hour session each week under supervision.
Participants were an average 77 years old, and 61% were men. The exercise group were fitter, but had marginally higher Alzheimer's disease assessment scores compared with the rest.
More than a third of the people invited to take part in the study declined, and 60% of the participants were men, which is unusual in dementia studies because more women than men have the condition. "We did not pre-specify a value for a negative effect, but the average effect observed was smaller than our pre-specified superiority target of 2.45 ADAS-cog points". Activities of daily living, number of falls, and quality of life were also assessed. What also isn't clear is whether other, perhaps longer, interventions would have made a difference.
It's important to note this does not change what we know about exercise's ability to protect against dementia.
"It also may be that vigorous exercise is not the solution".
The disappointing results are a setback for researchers, who had hoped an exercise programme might improve people's ability to carry out everyday tasks such as washing and dressing.
Currently, as a dementia therapy that does not involve medication, the NHS recommends group cognitive stimulation therapy classes, where sufferers undertake exercises created to improve their memory, problem-solving skills and language ability. And the number of people who declined to participate in the study was high; more men than women participated, even though dementia is more common in women in western Europe.
The researchers pointed to some trial limitations.
The physical fitness of the exercise group improved over the first 6 weeks of the exercise programme, as measured by the 6-minute-walk test.
The researchers reported having no conflicts of interest.