In the western world, current demographic trends are characterized by declining birth and death rates. This combination of trends has inevitably produced an increase in the aging of the population, being the sector of the elderly population the one that has suffered a bigger growth rate in recent decades in the West (not forgetting that the population of many countries of the so-called third world are beginning to increase their life expectancy). This increase in life expectancy leads to an increase in morbidity, and as a result, a high percentage of elderly people are currently facing various physical and cognitive impairments associated with age. These deteriorations require additional resources in health systems that aim to prevent disability and functional deterioration, rather than the number of years achieved. Therefore, one of the main objectives of current health systems is prevention. Within the strategies for prevention, diagnosis and treatment in geriatrics, it includes the detection of the so-called geriatric syndromes because they constitute a frequent cause of functional or social disability. The geriatric syndromes were defined by Kane in 1989, in his book Essentials of Clinical Geriatrics, as geriatric problems and constitute the following: immobility, instability and falls, urinary and fecal incontinence, dementia and acute confusional syndrome, infections, malnutrition, alterations in sight and ear, constipation, depression / insomnia, iatrogenia, immunodeficiencies, and impotence. This thesis studies one of the most common problems in the population of older people and one of the so-called geriatric syndromes: insomnia. Sleep disorders are common in older people and insomnia represents the most frequent of them. These disorders are not inherent in the aging process and when they do appear, they can significantly reduce the quality of life, trigger various pathologies, and increase the risk of frailty. Fragility is currently considered an independent geriatric syndrome that represents the decrease of the physiological reserves of the elderly with an increase in their vulnerability to adverse health outcomes [Romero, 2010]. Linda Fried et al. published in 2001 one of the most accepted definitions of fragility. For this, they established five criteria: involuntary weight loss of more than 5 kilos or 5% of body weight during the previous year, fatigue or low resistance to small efforts, decrease in muscle strength (evaluated with a dynamometer), reduced physical activity and slow speed for the march. The presence of three of the above criteria constitutes the definition of frailty and among them, the most important predictors of frailty appear to be weight loss and fatigue or exhaustion [Xue, 2008]. The fatigue and exhaustion caused by the lack of sleep would therefore be associated with the risk of fragility and consequent disability. On the other hand, the detection of sleep disorders is essential to detect early other health problems considered in turn geriatric syndromes such as depression, often associated with insomnia and risk of frailty. Sleep disorders especially affect institutionalized people and, however, there are few studies focused on this population. That is why an institutionalized geriatric population has been chosen as the object of study in this thesis. As a preliminary study, the thesis makes a review of all current methods of sleep study. These methods can be classified into objective methods (such as polysomnography or actigraphy) and subjective methods (such as diaries and sleep questionnaires). From all the methods studied, three have been chosen for cross-validation in two populations of 99 and 62 people institutionalized in six different centers in the province of Valencia. The subjective quality of sleep was studied in the first population, while the objective quality of sleep was also studied in the second population. Specifically, from the previous study of sleep assessment methods, actigraphy was chosen as the objective method. All participants in the second study carried a wrist actigraph during a week that recorded their sleep parameters objectively. As subjective methods, the sleep questionnaires of Athens and Oviedo were selected. All the participants completed both questionnaires, thus registering the subjective perception of their own sleep. In addition, the study included other demographic variables (gender, age, marital status), medical variables (prescribed drugs, illnesses, etc.), psycho-geriatric variables (Barthel, Tinetti, and Mini-Mental scales; and the Charlson morbidity index), and physical (analysis of blood and saliva, including the concentration of cortisol). This study aims to obtain information about the relationship between all these variables, with special interest in determining whether there is a significant correlation between the subjective perception of the sleep and the objective measurement of it. The study also aims to identify possible markers for the early detection of sleep disorders. The results obtained are, on the one hand, an updated classification of the sleep detection methods, as well as a historical compilation and comparison of the sleep questionnaires. On the other hand, the statistical analysis of the study carried out on the sample of participants yields several interesting results. First, a significant correlation was found between the scores obtained in the Athens questionnaire and the subscale 1 of the Oviedo questionnaire (rho = -0.51 p <0.01, Spearman correlation test), subscale 2 of the Oviedo questionnaire (rho = 0.62 p <0.01, Spearman's correlation test) and the total score in the Oviedo questionnaire (rho = 0.57 p <0.01, Spearman's correlation test). On the contrary, no significant correlation was found between the score in the Athens questionnaire and the subscales of the Oviedo questionnaire related to hypersomnia or the use of sleep aids or the presence of adverse events during sleep. The age and the Charlson comorbidity index were identified as significantly correlated (rho = 0.32, p <0.05, Spearman correlation test), but none of them was significantly correlated with the scores of the Athens questionnaire or any of the subscales of the Oviedo questionnaire. No significant correlation was found between age and sleep problems. However, a significant correlation was found (p <0.05, Spearman test) between the number of drugs taken daily and the quality of sleep measured both by the Athens scale and by the Oviedo questionnaire. In relation to sleep assessment by actigraphy, sleep efficiency in this population reached the average value of 95.31% ± 2.57%; the total sleep time was 321.92 ± 105.84 minutes; the average number of awakenings during sleep was 5.47 ± 3.32; and the average duration of awakenings of 2.59 ± 1.01 minutes. A significant and inverse correlation was found between the number of awakenings during the night measured with actigraphy or the total awake time during the night and the total score of the Oviedo questionnaire (rho = -0.32, p <0.05, and rho= -0.37, p <0.05, respectively). No significant effects were observed between the other variables collected with actigraphy and the other subscales of Oviedo or with the Athens questionnaire. There was, however, a significant correlation between the diagnosis of insomnia with the Athens questionnaire and the number of awakenings during the night and with the time awake during the night (rho = -0.34, p <0.05, and rho= -0.38, p <0.05, respectively). We also identified a significant and inverse correlation between age and total sleep time recorded by the actigraph (rho = -0.33, p <0.05, Spearman's correlation test) or bed time during the night (rho= -0.32, p <0.05, Spearman's correlation test). Finally, the concentrations (mg / dL) of cortisol in blood and saliva collected during the morning were studied in relation to sleep variables. No significant correlation was observed between the concentration of cortisol in saliva and any variable of sleep measured by actigraphy, but a significant correlation of the cortisol concentration was found in both blood and saliva with some subscales of the Oviedo questionnaire. On the other hand, a significant correlation was observed between blood cortisol levels and a short sleep time (<4 hours) measured with actigraphy. In particular, those participants who slept on average less than 4 hours had high concentrations of cortisol in the blood plasma (rho = 0.56, p <0.01, Spearman's correlation test). No correlations were observed with the Charlson index or with the number of medications administered daily. In conclusion, in addition to the correlations identified in the study, our results show the existence of possible biomarkers for the detection of sleep disorders. The relationships found in this study are a starting point to know the causes and mechanisms that produces a sleep disorder. An in-depth knowledge of these mechanisms will allow the development of medical and nursing interventions aimed at treating, preventing and even reversing sleep disorders, reducing the risk of frailty, and improving in this way the quality of the elderly people.
