Skip to main content
Fig. 3 | BMC Geriatrics

Fig. 3

From: The identification and prediction of frailty based on Bayesian network analysis in a community-dwelling older population

Fig. 3

Risk reasoning of Bayesian network model for frailty identification and frailty transition. A: Risk reasoning of Bayesian network model for frailty identification; B: Risk reasoning of Bayesian network model for frailty transition. Age: young = ≤ 71 years old, old = > 71 years old; IADL: instrumental activities of daily living, normal: IADL = 8, impaired: IADL < 8; balance capacity: normal: timed up and go test ≤ 10.33s, impaired: timed up and go test > 10.33s; malnutrition: yes: MNA-SF score < 12, no: MNA-SF score ≥ 12; anxiety: yes: GAD-7 > 4, no: GAD-7 ≤ 4; creatine: low: <66µmol/L, high: ≥66µmol/L; social support: normal: social support rating scale ≥ 27, social support rating scale < 27. Hospitalization: hospital admissions within one year; frailty transition: stable/improved or impaired according to the change of the Fried phenotype score; ADL/IADL: basic/instrumental activities of daily living; the change of the ADL/IADL score (14 scores in total) was calculated by baseline scores minus follow-up scores, subjects who received positive scores were considered as “ADL/IADL decline”; malnutrition: yes: MNA-SF score < 12, no: MNA-SF score ≥ 12; baseline DBP: low: <73 mm Hg, medium: 73 mm Hg ≤ DBP ≤ 87 mm Hg, high: > 87 mm Hg; baseline HDL: low: <0.99 mmol/L, medium: 0.99 mmol/L ≤ HDL ≤ 1.42 mmol/L, high: >1.42 mmol/L

Back to article page