Survival and early recourse to care for dementia: A population based study

A large proportion of dementia cases are still undiagnosed. Although early dementia care has been hypothesized to benefit both patients and families, evidence-based benefits are lacking. Thus, investigating the benefits for newly demented persons according to their recourse to care in the "real life" appears critical.

We examined the relation between initial care recourse care and demented individuals' survival in a large cohort of incident dementia cases screened in a prospective population-based cohort, the Three-City Study. We assessed recourse to care for cognitive complaint at the early beginning of dementia when incident cases were screened. We classified patients in three categories: no care recourse, general practitioner consultation or specialist consultation. We used proportional hazard regression models to test the association between recourse to care and mortality, adjusting on socio-demographical and clinical characteristics.

Two hundred and fifty-three incident dementia participants were screened at the 2 year or 4 year follow-up. One third of the incident demented individuals had not consulted a physician for cognitive problems. Eighty-six (34.0%) individuals had reported a cognitive problem only to their general practitioner (GP) and 80 (31.6%) had consulted a specialist. Mean duration of follow-up after incident dementia was 5.1 years, during which 146 participants died. After adjustment on potential confounders, participants who had consulted a specialist early in the disease course presented a poorer survival than those who did not consult any physician (hazard ratio = 1.64, 95% confidence interval 1.03-2.62). There was a trend but no significant differential survival profile between participants who complained to their GP and those without any care recourse.

Neither recourse to a specialist nor recourse to GP improve survival of new dementia cases. Those who had consulted a specialist early in the disease course even reported a worse life expectancy than those who did not.