Does gum disease raise the risk of dementia? What a ten-year Japanese study found

The research came out of the Hisayama cohort run by a team at Kyushu University, and was published in the Journal of Clinical Periodontology in August 2026. A total of 1,396 people aged 60 and over, none of whom had dementia at the start, were followed for ten years. Gum status was measured both at the start and in the fifth year; three measures were used: mean pocket depth, attachment loss and the percentage of sites that bled on probing.

By the end of the ten years, 267 of the 1,396 people had developed dementia. Of these, 194 were of the Alzheimer's type and 51 were vascular dementia. So roughly a fifth of the group was diagnosed within that period; the study was already following an elderly community.

For each of the three measures, participants were divided into four groups from best to worst. Comparing the worst group with the best, the risk of dementia came out at 1.72 times for pocket depth (95% confidence interval 1.18 to 2.46), 1.59 times for attachment loss (1.07 to 2.38) and 1.56 times for bleeding (1.09 to 2.22). For Alzheimer's type dementia, a significant association was seen only with the bleeding measure. For vascular dementia there was a trend for pocket depth, but that result did not reach statistical significance.

These three measures are not things the patient feels; they are things the dentist measures. Pocket depth is the gap between gum and tooth, in millimetres. Attachment loss is how much of the connective tissue holding the tooth in the jaw has gone. Bleeding on probing is the percentage of sites that bleed when touched during the examination, that is, how active the inflammation is at that moment. Bleeding you notice at home may be a rough sign of this, but it is not the same thing.

"1.72 times" is a relative figure, and read on its own it looks bigger than it is. A person's ten-year probability of dementia varies with their age, their family and their other conditions; this study does not let you recalculate that probability from your gums. What it says is more modest: within the same age group, a diagnosis was seen more often in those with poor gum measurements.

Nor is this a single study. An umbrella review published in 2026, pooling the meta-analyses on the subject, found the risk of mild cognitive impairment 1.65 times higher in people with gum disease, the risk of Alzheimer's 1.53 times higher and the risk of dementia 1.54 times higher. The same review placed the evidence in the weakest class (Class IV) and marked its certainty as "very low" on the GRADE scale. So the direction is consistent and the confidence is low.

For comparison: the 2024 Lancet Commission report on dementia prevention says that around 45% of cases worldwide are linked to 14 modifiable factors. Among those 14 are hearing loss, blood pressure, diabetes, smoking, inactivity, social isolation and air pollution. Gum disease is not on that list. While the literature examining the link keeps growing, the field's most widely used prevention list does not yet count it as an established factor.

How the link is made is not known either. There are proposed explanations: the mark long-running inflammation leaves across the body, oral bacteria and their products entering the bloodstream, difficulty chewing disrupting nutrition, and shared causes such as smoking and diabetes that affect both sides. The reverse is possible too: the early phase of dementia makes brushing the teeth and keeping appointments harder, so poor gums may be not the outcome itself but an early warning of it. The study cannot separate these possibilities from one another.

The distinction below shows the situations where this news changes something in practice, and the ones where it does not. None of it stands in for a personal assessment.

The three measures the study used are also the three measures of a routine gum examination. Knowing what is done at the appointment makes the chart you are given easier to read.

1. 1

   Probing

   A slim, blunt-tipped instrument marked in millimetres is placed in the gap between tooth and gum, and the depth is measured. Several points are checked around each tooth. In healthy sites this gap is shallow; its deepening means the inflammation has driven the tissue down further.

2. 2

   Recording the bleeding

   During the same probing, which points bleed is marked down and written as a percentage. This figure shows how active the inflammation is that day, and it is the measure that changes fastest after treatment.

3. 3

   Measuring the recession

   How far the gum margin has come down the root is measured. Added together with the pocket depth, it gives the attachment loss, that is, how much of the tissue holding the tooth has gone. This was the study's second measure.

4. 4

   The X-ray

   The pocket measurement describes the soft tissue; the film shows how far up the bone comes. The two are read together, because two pockets of the same depth can have different amounts of bone beneath them.

5. 5

   Treatment and measuring again

   Cleaning the calculus and the root surfaces can be spread over a few appointments. The same measurements are then repeated; the aim is for the bleeding to subside and the deep pockets to become shallower. In a course of treatment with no repeat measurement, what the outcome was is simply not known.

6. 6

   Setting the maintenance interval

   Gum disease is not counted as finished; it is followed. The interval is set for the individual and the measurements are repeated at every check-up. When that interval is not kept to, the ground gained is usually given back.

Most of what can be done after reading this is what is already advised, independently of the dementia literature. The order follows the strength of the evidence.

The gap between the study itself and the news about it is the most misunderstood thing here. We have gathered it under five headings.

The sequence below describes the general course after gum treatment that does not need surgery. It varies from person to person.

1. The first few days

   There can be tenderness after the cleaning, and sensitivity to hot and cold. The gum margin may bleed more easily in the first few days; this usually eases as the inflammation subsides.

2. Between week two and week four

   Less bleeding when brushing is the expected course. As the gum recedes, root surface that was previously covered by the pocket can become exposed; this is not deterioration, it is inflamed tissue settling down.

3. Between week six and week eight

   The measurements are repeated. What is looked for is a drop in the bleeding percentage and shallower deep pockets. If some sites have not responded, further treatment is discussed at this stage.

4. Afterwards

   The follow-up interval is set for the individual, and the same measurements are taken at every check-up. Gum disease is not a past event; it is a condition that is monitored.

You will not find figures here. These are the headings to look at when reading a quote:

How many sessions

Whether the whole mouth is cleaned in a single session or area by area. The number of sessions sets both the time and the total.

Whether surgery is involved

Some deep pockets may not respond to closed cleaning. If the possibility of surgery is discussed from the start, the quote is read accordingly.

Whether measuring again is included

What shows the outcome of the treatment is the measurement taken again afterwards. If it does not appear in the quote, the review appointment needs discussing separately.

The maintenance interval

Gum treatment is not a one-off job. How often the follow-up appointments come and what they cover should be clear from the start.