Frailty in an elderly general population. (Doctoral thesis)

Frailty is a term used to describe older people who are more vulnerable to stressors and therefore have a higher risk of death and disability. Frailty is not an irreversible condition and can be reverted with intervention such as physical exercise and nutritional support. Therefore, it can be argued that should be detected early. For this purpose, several frailty scores based on different frailty concepts have been developed. However, to date, none of them is recognised as the "gold standard". One aspect is to diagnose frailty but the other is to prevent the development of this condition. Understanding frailty and its determinants is crucial for prevention and treatment.
Most frailty scores have been studied in their association with mortality. However there is a gap in the literature concerning their agreement and external validation and discriminative ability.
Diabetes is known an important determinant of frailty and in addition, they share pathophysiological mechanisms. Frailty is not a static condition and tend to progress with age. However, some individuals can have different accelerated frailty trajectories, and they can even change the trajectory over time. The effect of diabetes over frailty trajectories is scarcely investigated to date.
The main objectives of this PhD thesis were to compare the current operational definitions of frailty and their instruments, through the evaluation of agreement among frailty scores and their predictive/discriminative ability as well as to study the association of between diabetes related variables and frailty progression.
This PhD thesis provides a direct comparison of the most comprehensive list of frailty scores examined to date, with state-of-the art and reproducible methodology, in a well-characterised cohort of the elderly general population.
In the first chapter, a general overview, objectives and hypotheses of the thesis are presented. Important basic concepts and methods that have been applied throughout the PhD work are described. Also, I describe the study population, the English Longitudinal Study of Ageing (ELSA study).
In the second chapter, the study entitled: “Agreement between 35 Published Frailty Scores in the General Population” is presented as Study I. In this article, I studied the cross-sectional agreement between 35 frailty scores in the ELSA study. I found marked heterogeneity in the degree to which the various scores may over/underestimate frailty and in the agreement on the identification of the same individuals as frail. I concluded that most of scores cannot be assumed to be interchangeable, and that consequently research results based on different scores cannot be compared, pooled or summarised directly.
In the third chapter, the study entitled: “Comparative analysis of the association between 35 frailty scores and cardiovascular events, cancer and total mortality in an elderly general population in England: an observational study” is presented as Study II. This study analyses the prospective association and predictive ability of 35 frailty scores in the ELSA study for three relevant outcomes in elderly population: mortality, cardiovascular disease and cancer. I demonstrated that all frailty scores were associated with future mortality and that some of them were also associated with later cardiovascular events. However, no relationship with cancer was observed. In addition, the results of this study showed that multidimensional frailty scores may have a stronger and more stable association with mortality and incidence of cardiovascular events. Despite significant associations of frailty scores with mortality outcomes, I found that the added discriminative ability of frailty scores to chronological age may be limited.
In the fourth chapter, the study entitled "Prospective association of baseline diabetes related variables and frailty trajectories in an elderly general population" is presented as Study III. I studied the baseline diagnosis of diabetes, baseline fasting plasma glucose, and HbA1c as determinants of frailty trajectories calculated with the three best-performing frailty scores identified in our two previous studies. I found that with 10 years of follow-up, baseline diagnosis of diabetes and baseline levels of HbA1c were associated with frailty trajectories, but not baseline fasting plasma glucose. I concluded that diabetes can be associated with frailty trajectories not only because of common pathophysiological mechanisms, but also because of chronic complications related to diabetes.
These three studies were based on the analysis of the same population: the ELSA study and included a literature review for identifying frailty scores, a data analysis to calculate scores and a multiple imputation technique to deal with missing data. They follow one another in a logical order of analysis to give answers to the research questions.
Finally, in the fifth chapter, I discuss our results and their relevance, particularly in the way this thesis contributes to a better understanding of the concept of frailty and its contribution to knowledge in this field so far. In addition, I discussed strengths and weaknesses of the analyses presented in this thesis, and I suggest some recommendations derived from the findings of the studies for clinicians and researchers suggesting future directions for research.