What a relative risk of 1.85 actually tells you
A relative risk compares the chance of developing dementia in people with a sleep problem to the chance in people without it. A value of 1.00 means no difference; 1.85 means the risk is 85% higher. These are not personal predictions — they describe groups over years of follow-up, and the true size of the effect can shift with age, health status, and how sleep was measured.
The comparison across subtypes matters because the same sleep complaint can point in different directions depending on which dementia is being counted. In the largest synthesis, 76 cohort studies were pooled, and the pattern was not uniform: some sleep problems leaned toward vascular dementia, others toward Alzheimer's disease [1].
Daytime sleepiness leans toward vascular dementia, not Alzheimer's
Excessive daytime sleepiness — feeling sleepy during the day despite time in bed — was associated with a 1.85-fold risk of vascular dementia and a 1.41-fold risk of all-cause dementia in the pooled analysis [1]. That is a larger signal than insomnia showed for dementia overall (1.13) [1].
A separate 12-year cohort of adults aged 65 and over found the same direction: daytime sleepiness raised the risk of all-cause dementia (1.21) and dementia with a vascular component (1.58), but not Alzheimer's disease [5]. The two studies used different subtype labels and different populations, so the exact numbers are not directly interchangeable, but both point to a vascular link rather than an Alzheimer's one.
One caution sits beside this finding: in that cohort, sleepiness was assessed once at baseline by self-report, and the authors note that mechanisms and whether treating sleepiness would prevent dementia remain unresolved [5].
Long sleep is the stronger signal for Alzheimer's disease
Long sleep duration was associated with a 1.66-fold risk of Alzheimer's disease, a 1.43-fold risk of all-cause dementia, and a 1.23-fold risk of cognitive decline in the pooled analysis [1]. That makes long sleep the sleep measure most clearly tied to Alzheimer's in these data, in contrast to daytime sleepiness, which was not associated with Alzheimer's in the cohort study [5].
Short sleep was also a risk factor in the pooled analysis, with a relative risk of 1.27 for dementia [1]. So both extremes of sleep duration carried elevated risk, but the subtype pattern differed: long sleep had the stronger Alzheimer's signal, while the pooled estimates for short sleep were reported for dementia overall rather than broken out by subtype [1].
An earlier study of older adults without dementia reported that long sleep duration, but not short sleep duration, was associated with dementia mortality, and it also examined dementia subtypes [3]. That is consistent with long sleep being the more distinctive marker, though the outcome there was death with dementia rather than incident dementia.
Insomnia is a smaller and less consistent signal
Insomnia was associated with a 1.13-fold risk of dementia in the pooled analysis — a real but modest elevation compared with the sleepiness and long-sleep estimates [1]. In the 12-year cohort, insomnia complaints showed no association with all-cause dementia, Alzheimer's disease, or dementia with a vascular component, whether considered alone or combined with daytime sleepiness [5].
The gap between these results may reflect how insomnia was defined and measured. The pooled analysis combined many cohort studies with differing assessments [1], while the cohort used a structured interview and self-questionnaire at a single baseline [5]. Neither result establishes that insomnia causes dementia; the pooled estimate is an association across studies, and the cohort authors describe their findings as suggesting a role for sleepiness and hypnotics rather than for insomnia [1][5].
Sources used in this answer
Sleep disorders and the risk of cognitive decline or dementia: an updated systematic review and meta-analysis of longitudinal studies
In a meta-analysis of 76 cohort studies, insomnia was associated with a 1.13 relative risk of dementia, short sleep with 1.27, long sleep with 1.23 for cognitive decline, 1.43 for all-cause dementia, and 1.66 for Alzheimer's disease, and excessive daytime sleepiness with 1.85 for vascular dementia, 1.41 for all-cause dementia, and 1.37 for cognitive decline.
Association between physical activity over a 10-year period and current insomnia symptoms, sleep duration and daytime sleepiness: a European population-based study
In a European population cohort, persistently active adults were less likely than persistently inactive adults to report difficulty initiating sleep (OR 0.60), short sleep of 6 hours or less (OR 0.71), and long sleep of 9 hours or more (OR 0.53), while daytime sleepiness was not related to activity status.
Long sleep duration in elders without dementia increases risk of dementia mortality (NEDICES)
In older adults without dementia, long sleep duration, but not short sleep duration, was associated with increased dementia mortality, and dementia subtypes were examined.
0285 Excessive daytime sleepiness with long sleep duration increases myocardial infarction risk
In UK Biobank participants, excessive daytime sleepiness with long sleep of 9 hours or more was associated with a 91% higher incidence of myocardial infarction, while sleepiness with normal or short sleep was not.
Complaints of daytime sleepiness, insomnia, hypnotic use, and risk of dementia: a prospective cohort study in the elderly
In a 12-year cohort of adults aged 65 and over, excessive daytime sleepiness was associated with higher risk of all-cause dementia (HR 1.21) and dementia with a vascular component (HR 1.58) but not Alzheimer's disease, while insomnia complaints showed no association with any dementia outcome.