Longitudinal dynamics of frailty in older people: Trajectories, transitions and associated clinical factors.

BACKGROUND: Frailty is a dynamic geriatric syndrome characterized by progressive deterioration related to age. OBJECTIVE: To describe the trajectories and estimate the transition probabilities between frailty phenotypes and identify clinical variables and drugs associated with the transition to more severe phenotypes over four years of follow-up in a cohort of elderly people living in the community. STUDY DESIGN: Secondary analysis of data from the study "Cohort of Obesity, Sarcopenia and Frailty of Older Mexican Adults". The baseline sample included adults ≥ 60 years, with annual monitoring for four years. MAIN OUTCOME MEASURES: Frailty phenotype, chronic diseases, multimorbidity, medications, polypharmacy, and sociodemographic variables. The trajectories were identified using the optimal alignment algorithm, and the longitudinal patterns of fragility were analyzed with hierarchical clustering techniques. The annual transition probabilities were estimated with a nested Markov model, and the factors associated with the transitions were identified using ordinal multilevel mixed-effects logistic regression models. RESULTS: The initial cohort included 973 patients. Four frailty trajectories were identified. The transitions between frailty phenotypes were asymmetric, and the lowest annual rate of occurrence was improvement. The effect of time (follow-up), each additional year of a person, being female, suffering from multimorbidity (MM) and receiving antihypertensive, psycholeptic and/or psychoanaleptic, antiparkinsonian, or antivertiginous drugs increased the probability of moving toward a more severe state of frailty; a higher level of education and having a partner conferred a protective effect. CONCLUSION: Early identification of the prefrailty condition offers a window of opportunity to reverse it or prevent further deterioration.