Moving more, sitting less, and getting enough sleep can reduce the risk of developing dementia.

A systematic review and meta-analysis published in PLOS One summarized data from prospective cohort studies on how three groups of daily habits—physical activity, sedentary behavior, and sleep duration—are associated with the risk of newly diagnosed dementia in community-dwelling adults. The authors specifically included only studies with follow-up to examine future dementia risk, rather than the current situation. The article was published on April 8, 2026.

The search strategy was broad. The team searched MEDLINE, EMBASE, PsycInfo, CINAHL, and SPORTDiscus for studies, covering publications from the earliest available years through August 2025. There were no language restrictions, but gray literature was excluded. The review included only studies with participants aged 35 years and older, without dementia at baseline, with at least one year of follow-up, and with valid behavioral and outcome measures.

The analysis focused on three behavioral indicators. For physical activity, the authors used regular activity as a guideline, consistent with national recommendations. For sedentary behavior, the threshold was defined as 8 hours of sitting per day or more. For sleep, short sleep (less than 7 hours) was compared with long sleep (more than 8 hours), with a reference range of 7-8 hours per night. The analysis was performed using a random-effects model, with separate subanalyses based on age and duration of observation.

This is important because the study looked not at a single factor, but at a behavioral profile that can actually be modified. The authors emphasize that dementia has a long preclinical period, and pharmacological approaches are still limited, so studying modifiable behaviors remains one of the key strategies for preventing or delaying the disease.

Research component	What was included in the analysis
Type of work	Systematic review and meta-analysis of cohort studies
Population	Adults 35 years and older living in the community
Databases	MEDLINE, EMBASE, PsycInfo, CINAHL, SPORTDiscus
Search period	Until August 2025
Main factors	Physical activity, sedentary time, sleep duration
Sleep standard	7-8 hours per night
Sedentary time threshold	8 hours a day or more

Source of the table. [1]

What the physical activity analysis showed

For physical activity, 49 studies were included in the quantitative synthesis. These included a total of 2,855,529 participants, of whom 97,167 were diagnosed with dementia during follow-up. The average age across these studies was 67.3 years, and the average follow-up was 11.6 years. This is the largest block of the meta-analysis.

The main result here was in favor of physical activity. The overall risk of dementia in people who regularly engaged in physical activity was lower: the pooled risk was 0.75, or approximately 25% lower than in less active participants. Moreover, 25 of the 49 studies, or 51.0%, already showed a statistically significant risk reduction.

But this result comes with an important caveat. Heterogeneity between studies was high, with I² = 92.5%. This means that the effect size varied significantly across studies. Subanalyses by age and follow-up duration slightly reduced the variability, but did not eliminate it completely. The authors explicitly state that other unaccounted sources of variation remain.

In practical terms, this means the following: the overall signal in favor of physical activity appears robust, but the precise magnitude of the effect cannot be considered the same across all ages, countries, and activity measurement methods. Yet, of the three factors studied, physical activity yielded the largest body of evidence and one of the most compelling protective signals.

Physical activity and dementia	Result
Number of studies	49
Number of participants	2,855,529
Cases of dementia	97 167
Middle age	67.3 years
Average duration of observation	11.6 years
Pooled risk	RR 0.75
Interpretation	More active people have a 25% lower risk of dementia
Heterogeneity	I² = 92.5%

Source of the table. [2]

What the sleep and sedentary behavior data showed

The sleep analysis included 17 studies with 1,344,170 participants, of whom 49,581 developed dementia during the follow-up period. The average age in these studies was 67.5 years, and the average follow-up was 12.1 years. All 17 studies compared short sleep (less than 7 hours) with long sleep (more than 8 hours), with a range of 7-8 hours used as the reference range.

The results were U-shaped. Both short and long sleep were associated with a higher risk of dementia. For short sleep, the combined risk was 1.18, while for long sleep, it was 1.28. In other words, sleep shorter than 7 hours was associated with an approximately 18% higher risk, and sleep longer than 8 hours was associated with an approximately 28% higher risk, compared to 7-8 hours. Moreover, the risk was higher for long sleep than for short sleep.

Sedentary behavior was less well-studied, but the results were fairly clear. Only three studies were included in the quantitative analysis, but together they covered 295,809 participants. People who spent eight or more hours per day sitting had a higher risk of dementia: the pooled coefficient was 1.27. Unlike the physical activity component, heterogeneity here was low, I² = 0%.

The practical implications of these two blocks appear quite clear. The best profile in this study is not only exercise but also relatively moderate sleep and less sedentary time. The authors specifically emphasize that 7-8 hours of sleep was associated with the lowest risk among the categories studied, and prolonged sitting appeared to be an independent adverse factor, not simply the downside of inactivity.

Sleep and sedentary behavior	Result
Sleep research	17
Participants in sleep analysis	1,344,170
Cases of dementia in sleep analysis	49,581
Short sleep < 7 hours	RR 1.18
Long sleep > 8 hours	RR 1.28
Studies on sedentary time	3
Participants in the sedentary time analysis	295 809
Sitting for 8 hours a day or more	RR 1.27

Source of the table. [3]

How to understand this and where are the boundaries of conclusions?

The authors interpret the results as evidence that regular physical activity, reduced sedentary time, and 7-8 hours of sleep may be modifiable factors associated with a lower risk of dementia. This is particularly important for public health, as these aren't rare or high-tech interventions, but rather everyday habits that can potentially be influenced at the individual, community, and policy levels.

However, the study does not prove causation. It is a pooled observational cohort study, not a randomized intervention. The authors explicitly warn that residual confounding and reverse causation cannot be completely ruled out, especially in studies with short follow-up periods, when early, undiagnosed changes may already be affecting sleep, mobility, and habits in some individuals.

There are also methodological limitations. Many studies assessed physical activity, sedentary behavior, and sleep by self-report, and did not consider sleep quality at all. Furthermore, the risk of bias for most studies on physical activity and sleep was moderate, while for most studies on sedentary behavior, it was high. Confounding factors and missing data were the most common problems.

Therefore, the correct conclusion is not "10,000 steps and 8 hours of sleep have been proven to prevent dementia," but a more cautious statement: a more active lifestyle, less sedentary time, and sleep in the mid-range of 7-8 hours are consistently associated with a lower future risk of dementia in large cohort data. This is a strong epidemiological signal, but not a definitive clinical verdict or a universal formula.

Limitation	Why is this important?
Observational design	Does not allow to prove causality
High heterogeneity in physical activity	The effect varied between studies
Self-report on behavior	Memory and measurement errors are possible
Sedentary time has been studied in only 3 studies	The evidence base here is thinner.
Most studies had moderate or severe risk of bias.	The results should be interpreted as associations, not as proven effects.

Source of the table. [4]

Source: Oye-Somefun A, Mirzadeh P, Gao-Kang J, Rotondi M, Kuk JL, Tamim H, Ardern CI. The Relationships between Physical Activity, Sedentary Behavior, Sleep, and Dementia: A Systematic Review and Meta-Analysis of Cohort Studies. PLOS One. Published April 8, 2026. DOI: 10.1371/journal.pone.0343621.