Background Although studies have demonstrated a link between poor oral health of older people and frailty, these are mostly cross-sectional or have limited oral health measures. We investigated the associations of objective and subjective measures of oral health with incident frailty and disability in older people over a three-year follow-up period.

Methods Data come from the British Regional Heart Study, a cohort comprising a socially and geographically representative sample of older British men, initiated in 1978–80. The participants underwent an examination at 71–92 years in 2010–12 including a dental examination for tooth count, and periodontal disease (response rate 68%). Self-reported oral health measures comprised overall self-rated oral health, dry mouth (or xerostomia) symptoms, sensitivity to hot/cold/sweet and difficulty eating. A cumulative measure of poor oral health combined absence of natural teeth (edentulism), fair/poor self-rated oral health, dryness of mouth, sensitivity to hot/cold/sweet, and difficulty eating. In 2014, 1,222 participants completed a follow-up postal questionnaire. Frailty was based on presence of at least three self-reported components of weight loss, low grip strength, feelings of exhaustion, slowness, or low physical activity. Disability was based on mobility limitations (difficulty walking 400m/going up or down stairs). Covariates included age, alcohol consumption, smoking, depression, social class, social interactions, history of diabetes or cardiovascular disease, use of medications with dry mouth as a side-effect, and interleukin-6 levels.

Results During the follow-up period, 99 (10%) men presented with incident frailty and 141 (15%) with incident disability. Fully adjusted logistic regression models showed that the risk of becoming frail was higher in participants who were edentulous (odds ratio (OR) = 1.84, 95% CI: = 1.03–3.28), those reporting at least one dry mouth symptom (OR = 1.76, 95% CI: 1.07–2.88), and with cumulative oral health problems (OR for an additional oral health problem was 1.36, 95% CI: 1.03–1.80; p for trend ≤ 0.0001). The risk of incident disability was greater in those with fewer teeth (<21 teeth) (OR = 1.93, 95% CI: 1.19–3.12), dry mouth symptoms (OR = 1.60, 95% CI: 1.06–2.43), cumulative oral health problems (OR for an additional oral health problem was 1.28, 95% CI: 1.01–1.62), and periodontal disease (OR = 1.88, 95% CI: 1.12–3.16) in fully adjusted models.

Conclusion Several oral health measures predict incident frailty and disability. Poor oral health could be an early marker of frailty and disability in older people.