Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Incidence of dementia


Differences in dementia incidence and prevalence in older people around the world are believed to be related to vascular risk. Because different studies use different diagnostic criteria in different populations, comparisons between them are of limited value in relating development of dementia to underlying health problems. Bandolier 48 looked at a comparison of different diagnostic criteria applied by the same people to a large cohort of older people with full clinical and neuropsychological examination. That study showed that the diagnosis of dementia in the same population ranged from 3% to 30%.

In order properly to examine issues of underlying health and their relation to dementia, different populations have to be examined using the same diagnostic criteria applied in the same way. A large UK MRC study has done that [1].


The study was carried out in five areas around the UK. Two (Gwynedd, Newcastle) had higher rates of angina, intermittent claudication, heart attacks, and stroke. Cambridgeshire and Oxford had lower rates, and Nottingham was intermediate. Population samples were drawn from health authority lists, with random selection to recruit at least 1,250 individuals in each centre in the age groups 65-74 years, and 75 years and above.

After an initial screening interview, there was a more intensive assessment of a subsample of the population. After two years, there was a rescreen of individuals not selected for assessment at baseline, followed by a further assessment sample. Interviewers were trained to use standard formats, with considerable effort made to ensure ongoing comparability locally and across all sites. The interview lasted about an hour and contained basic information plus a variety of instruments related to dementia. Diagnosis of dementia was based on DSM-III, with clinical judgement for those unable to complete the interview.

Essentially the incidence rate was calculated as the proportion of people with incident dementia, divided by two years, and presented as a rate per 1,000 person-years.


As with other studies, incidence of dementia increased with age for women and men (Figure 1). Incidence was about 10 per 1,000 person-years (about 1% of the age group) in men and women between age 70 and 80 years, with rates increasing to about 60 per 1,000 person-years (about 6%) in those aged 85 years and older. There was no indication that rates did not continue to increase in the oldest old.

Figure 1: Incidence of dementia by age in women and men

Although individual sites differed considerably in rates of vascular disease, there was no indication of a lower incidence in sites with the lower rates of vascular disease (Figure 2). Extrapolation to the population of England and Wales indicated that there would be about 180,000 new dementia cases occurring each year.

Figure 2: Overall incidence of dementia in persons over 65 years in five UK areas with very different levels of vascular risk factors


This is a complex and detailed study that performed a number of sensitivity analyses and examined various issues of methodology. Its findings were broadly in line with others in Europe, Asia, and the USA. Its value was that it indicates that there is no apparent link between vascular health and dementia, and confirmed that rates do not level off or decline in the oldest old.

The findings have implications for planning of health services in the face of substantial increases of older people in the population now, and in the future, as well as the increasing proportion of older people in the population. It has interesting societal overtones as well, especially when newer prevention trials have only moderate preventative effects, with NNTs of about 90 over three years to prevent one case of cognitive impairment [2].


  1. 1 F Matthews et al. The incidence of dementia in England and Wales: findings from the five identical sites of the MRC CFA study. PLoS Medicine 2005 2: e193.
  2. K Yaffe et al. Effect of raloxifene on prevention of dementia and cognitive impairment in older women: the multiple outcomes of raloxifene evaluation (MORE) randomised trial. American Journal of Psychiatry 2005 162: 683-690.

previous or next story