We aimed to examine associations between diet quality, falls risk, physical function, physical activity and body composition.

Cross-sectional study.

Data collected from 171 men and women, aged 60–88 years old, as part of the Falls Risk and Osteoporosis Longitudinal Study.

Dietary Intake (Dietary Questionnaire for Epidemiological Studies Version 2 (DQES v2)), Falls Risk (FES-I, ABC, Berg Balance and Physiological Profile Assessment), Physical Function (SPPB), Physical Activity (PASE) and Body Composition (fat mass, lean mass, BMD, BMI, android/gynoid ratio) were ascertained. Diet quality was determined using two measures (Healthy Eating Index - HEI and Healthy Diet Indicator - HDI). One-way Analysis of Variance was used to compare mean scores between females and males and Pearson product-moment correlation coefficients were calculated to examine bivariate relationships.

Although females and males were analysed separately, the HDI-total score showed more associations that the HEI in both genders. The HDI showed, in females weak negative associations with BMI (r =−.21, p=.04), gynoid fat (r =−.20, p=.01), total fat mass (r =−.20, p=.02), with a weak positive association between HDI and percentage lean mass (r =.20, p=.03). Males showed positive associations between HDI and age (r =.30 p=.02) physical function (SPPB)(r =.26, p=.04), and subjective falls-risk (ABC) (r =.26, p=.03). In addition, in males, a negative association was found between HDI and FES-I (r =−.25, p=.04). The only measure that was significantly associated with the HEI-total score was the android/gynoid ratio in males (r =−.29, p=.04). When controlling for age, females demonstrated weak positive associations between gynoid (r =.19 p =.02), android (r =.19, p =.02) and total fat mass (r =.20 p =.02) as well as weak negative correlation with lean mass (r = 1.19, p =.03). Age also impacts on the FES-I (r =.29 p <.01) and ABC (r = −.23 p <.01).

The relationships between dietary quality and body composition, falls risk and physical function in older community dwelling, higher functioning adults appear to be gender specific. Better diet quality in females, is associated with lower BMI and fat mass, and higher lean mass, compared to males that are older and appear to have better physical function, are less likely to self-report falls risk, and have a better fat distribution i.e. a lower android/gynoid ratio have better diet quality. Furthermore, age is an important confounder and should be taken into consideration when assessing diet quality in older adults. In addition these gender and age differences may be clinically relevant and could aid in the delivery of targeted interventions.