The contribution of the 10/66 studies
Despite some excellent studies, dementia in LAMIC remains under-researched, particularly given its large, growing public health and societal impact. There are inequities in the increasing attention directed towards NCDs in LAMIC. Relatively little research, policy or practice is directed towards older adults. Premature mortality is given more attention than living with disability. According to the 2003 World Health Report Global Burden of Disease estimates, dementia contributed 11.2% of all years lived with disability among people aged 60 and over; more than stroke (9.5%), musculoskeletal disorders (8.9%), cardiovascular disease (5.0%) and all forms of cancer (2.4%). Our programme of population-based studies in ten developing countries provides a powerful resource for comparative descriptive research of prevalence, incidence, impact and cost. The main aim of the incidence phase is to investigate aetiology (a particularly neglected area). Findings from the west (for example, cardiovascular risk factors as risk factors for dementia) may, or may not generalise. Both exposure prevalence and effect sizes may differ between regions because of population-specific genetics, culture (behaviour and lifestyles) and physical environment. Other exposures, for example micronutrient deficiency, are rare in richer countries but highly prevalent and potentially of great significance in Latin America and China. Population attributable fractions will assist in determining priorities for future health promotion and primary care prevention in these regions. While cardiovascular disease and its risk factors are also relatively little studied in Latin America, two new initiatives will soon contribute relevant data; the PREVENCION study in Arequipa, Peru involves a comprehensive assessment of cardiovascular risk factors and outcomes among 1600 community residents aged 20–80, while the CARMELA study will provide similar data in younger samples (aged 25–64) from seven countries – Colombia, Argentina, Peru, Mexico, Ecuador, Chile and Venezuela. Our 10/66 population-based studies will neatly complement the CARMELA database by providing information from many of the same countries on those aged 65 and over.
Our one phase dementia diagnostic assessment has advantages over the two phase approach used in most previous dementia cohort studies . Attrition is marked between the first and second phase ; participants with probable dementia are particularly likely to refuse, to move away or to die, leading to informative censoring. The problem is compounded when no random sample of screen negatives is selected for second phase assessment with the tacit assumption of perfect sensitivity for the screening measure [104–106].
The 10/66 dementia diagnosis has been carefully validated across all of the cultures, and in each of the centres in which the population research is proceeding. While sensitivity (94%) and specificity (97% in high education controls and 94% in low education controls), against the gold standard of a local clinician's DSM IV diagnosis were both excellent, the false positive rate which varied between 1% and 10% across regions and levels of education can be expected to result in a higher prevalence of 10/66 dementia, compared with that of DSM IV dementia. DSMIV criteria are recognized to be relatively restrictive. Impairment is required in memory and other specified domains of function. Each of these must have progressed, and led to social or occupational impairment. The deficits should not be explained by delirium or other mental disorder. The aim with DSMIV (and other similar diagnostic criteria) was to define a progressive and relatively pervasive disorder with thresholds set high to maximize reliability between raters and centres. Clinically relevant dementia may therefore be prevalent beyond the confines of the narrowly defined DSM IV criterion. One of the few population-based studies to examine this issue directly, the Canadian Study of Health and Aging  reported a prevalence of 20.9% for those aged 65 and over according to clinical consensus compared with 13.7% according to DSM IV criterion. Mild cases, confirmed by clinicians, were selectively excluded by the DSM IV criterion. None of the DSM IV criteria are specifically required by the 10/66 probabilistic algorithm, which simply requires a profile of cognitive impairment on formal testing, informant reports of cognitive and functional decline, and findings on clinical interview that are consistent with a high probability of being a case. While the DSMIV algorithm identifies clear cut, severe and pervasive dementia cases, the 10/66 algorithm may be more relevant to establishing the true population burden of the dementia syndrome.
The catchment area sampling strategy enables us to foster links within each local community, improving response and facilitating possible follow-up. Community sensitisation proved to be essential to ensure a good response to the surveys. Prevalence estimates and other descriptive elements may not generalize beyond these and similar communities, but this is unlikely to lead to bias in estimates of association.
An action research program of this kind stands or falls upon its ability to inform and encourage policy development on the basis of the evidence accumulated through its activities. The 10/66 Group recently held a one week workshop at the Rockefeller Center in Bellagio to address the need to exploit fully the potential created by the 10/66 population based and intervention studies. Dissemination through a peer reviewed scientific journal is but one important element of this process. Our challenge is to use our findings to raise public awareness, stimulate local clinical training and practice, and influence social welfare and health care policy making at the national and international level. In these respects, our relationship with ADI (Alzheimer Disease International) is crucial. ADI is affiliated to the World Heath Organisation. The needs of people with dementia in developing countries is now a major priority for ADI, and our findings are disseminated on its website , at its conferences, and in its regular newsletters and World Alzheimer's Day Bulletins, distributed to its 76 member associations worldwide. In turn, the national Alzheimer's associations are able to use our materials for local publicity, and to influence national policy makers as part of their lobbying activities at government level. Other committed NGOs may have a key advocacy role to play, and we will need to work intersectorally to maximise dissemination and policy impact. Dementia is one of many health conditions in the developing world characterised by lack of awareness, stigma, limited help seeking, few services, and much unmet need. The evidence provided by research, disseminated actively by committed NGOs can be a powerful argument for change.
At the Bellagio meeting we devised a strategy for dissemination focusing upon: Identifying key stakeholders; designing and conducting local workshops; forging links with policymakers, understanding their preoccupations, and preparing policy briefings in each of the countries/regions where we are working; linking with the print and radio news media in developing countries, and understanding how its support may be elicited in raising awareness; exploiting links between 10/66 researchers and national Alzheimer's Associations; forming intersectoral links with other relevant NGOs e.g HelpAge International, Save the Children (cross-generational effects), Oxfam (economic impact of aged care, and poverty reduction)
All investigators in this programme are committed to establish a monitored public access fully anonymised file sharing archive, to maximize exploitation of the data resource. The Brazilian centre is making its own local arrangements. For all other centres, data will be forwarded to the coordinating centre for further cleaning and checking for compatibility with the uniform file format. The final version will be approved by the local centre, archived in a secure directory on the Institute of Psychiatry network, and shared with all centres participating in the archive. Proposals for multi-centre publications may be made by the coordinator, by centre PIs, or by external investigators. Access will not unreasonably be denied, and will be approved by a publications committee with two members from each centre. The 10/66 programme steering committee will give independent oversight. In the interests of transparency, all interviews, data entry files, algorithms, protocols, training materials and manuals can be downloaded from our intranet site. Applications for use of the data should be sent to email@example.com.