Ethical Aspects
The Ethics Research Committee of Federal University of Paraiba approved this research protocol (CAAE: 66122917.6.0000.5188), in accordance with the ethical standards of the national research committee, as well as with the 1964 Helsinki declaration and its later amendments. All participants gave a free written informed consent.
Research Scenario
This study was carried out in a capital city of Brazilian Northeast (João Pessoa, Paraíba, Brazil - Latitude: 07º 06' 54" S, Longitude: 34º 51' 47" W). This municipality has 800,323 inhabitants, a human development index at 0.763 and a per capita gross domestic product of R$23,169 (roughly US$ 6,400). Within the metropolitan region of João Pessoa, there are seven long-term care institutions, which assist an average of 50 elders per institution. Institutions are philanthropic and most of costs are covered by elders’ income (average US$ 250 per month).
Subjects and Study Design
Institutionalized elderly population in the metropolitan region of João Pessoa consisted of 398 individuals. In a pilot study, we detected that the response rate was around 40%, since many of the institutionalized individuals had seriously cognitive impairment. A design effect of 1.4 was calculated and the sample size of 191 was set as representative of non-to-middle cognitive impaired elderly.
A cross-sectional study was then conducted with 193 institutionalized elderly people living in seven different long-term care facilities located in a capital city of Brazilian Northeast. Inclusion criteria required elderly to assimilate the methodological tools and agree to participate in the survey. Initial screening of participants was achieved after subjective evaluation and assessment of cognitive status with regards to space-time orientation. A minimum of 13 points within the Mini-Mental State Examination was determined to consider the elderly able to answer the research questionnaires. In addition, subjects with chronic degenerative diseases (i.e. Azheimer disease and Parkinson disease) were not included within the study. Subjects answered the questionnaires after agreeing participate in the research and sign an informed consent.
Seven previously trained researchers took part in this survey. Training of researchers involved a theoretical exposition of all validated instruments, as well as a clinical experience to set the collection procedure. Concordance was set within the group of examiners as above 0.9.
Questionnaire and Variables
The following independents variables were included in the present study: socio-demographic characteristics (sex, age, education level, retirement and family visits) and data associated with general health (Performance of daily-living activities, Frailty status, Cognitive status, Nutritional status, Self-perception of oral health and Depression status). Health-Related Quality of Life of life – HRQoL was considered a dependent variable in this study. All data were collected using validated questionnaires, which were used to interview the subjects.
Performance of Daily-Living Activities (Katz scale)
The performance of daily-living activities was assessed using a six items questionnaire that measured the individual’s performance in self-care activities, including the following domains: 1) feeding; 2) sphincter control; 3) transference; 4) personal hygiene and use of the toilet; 5) dressing ability; 6) taking a shower [14]. Each dependence score was considered one point. In this survey, participants who were dependent of two or more functions were considered dependent.
Frailty Status
Frailty status was evaluated using a self-reported instrument validated by Nunes et al. [15] and adapted from the original instrument proposed by Fried et al. [16]. According to this instrument, five criteria are evaluated: non-intentional weight loss and muscle force reduction, poor endurance and energy, slowness and low physical activity level. In this survey, participants were categorized as frail when they have three or more positive scores for frailty. Muscle force was evaluated using a handgrip dynamometer; however, the handgrip force was not included in the statistical model
Cognitive Status
The Mini-Mental State Examination (MMSE) was used to measure the cognitive impairment and used as a screening for dementia [17]. The MMSE is composed of typical questions grouped into seven categories, each of which aims to evaluate specific cognitive functions: orientation to space-time, word registration, attention and calculation, word recall, language, and visual construction [18]. In this survey, elderly were included in two categories: with no cognitive impairment (≥ 21 points) or with cognitive impairment (< 21 points).
Nutritional Status
The Mini Nutritional Assessment Short Form (MNA-SF) is a well-validated technique for evaluating the risk of malnutrition among elders without the help of a dietician or nutritionist. It is based on anthropometric measurements, a global assessment of general health status and a subjective assessment (self-perception) of health and nutrition. Higher score indicated a more satisfactory state of nutrition [19]. In this survey, elderly were included in two categories: normal (≥ 8 points) or undernourished (< 8 points).
Self-Perceived Oral Health
The Geriatric Oral Health Assessment Index (GOHAI) was used to assess the self-perception on elderly’s oral health. For this, a 12 items questionnaire analyzed the physical function, psychosocial function and pain or discomfort. It pays special attention to problems related to food ingestion, which are addressed by at least one item in all three dimensions of the index: ‘trouble biting or chewing food’ (functional limitation), ‘discomfort when eating’ (pain and discomfort), ‘uncomfortable eating in front of people’ (psychological impacts), and ‘limit kinds or amounts of food’ (behavioral impacts) [20]. Voluntaries answered questions as never (score 3), sometimes (score 2) and always (score 1). In this survey, elders were included in two categories: good perception (≥ 34 points) or bad perception (< 34 points).
Depression Status
The Geriatric Depression Scale (GDS) was used to assess depression, using a 15 items depression scale. It was developed to exclude the effects of non-specific somatic symptoms such as anorexia and insomnia, which are frequently observed among elderly populations [21,22]. Each item can have 2 answers (yes or no). The highest possible score is 15, which indicates the most severe depressive state [23]. In this survey, participants were included in two categories: without depression (≤ 5 domains) or suggestive of depression (> 5 domains).
Health-Related Quality of Life (HRQoL)
HRQoL was measured using the short form instrument, that consists of 12 items measuring the following eight concepts: physical functioning; role physical due to physical problems; role emotional due to limitations in emotional health; mental health; bodily pain; general health; vitality; and social functioning, which can be combined into two sum scales, physical and mental sum scales, that reflect physical and mental health, respectively [24].
In this survey, participants were included in two categories, according to the median value found in this study: bad HRQoL (< 62 points) or good HRQoL (≥ 62 points).
Theoretical-Conceptual Model
A theoretical-conceptual model was designed for this study (Figure 1), in order to determine factors related to Health-Related Quality of Life (HRQoL) of institutionalized elderly involved in this survey. Block 1 included distal independent variables (sex, age, education level, retirement and family visits). Block 2 included proximal independent variables related to general health (Performance of daily-living activities, Frailty status, Cognitive status and Nutritional status, Self-perceived oral health and Geriatric Depression Scale).
Statistical Analysis
A descriptive analysis was conducted to check absolute and relative distributions, as well as to calculate means, medians and standard deviations of data. Data were analyzed using IBM Statistical Package for Social Sciences software (IBM SPSS, v. 20, Chicago, IL). The independent variables were evaluated with regards their association with HRQoL through the use of two statistical regression models (multiple linear and multiple binary logistic), as shown in Figure 1. For multiple linear regression model, variables were used in their continuous form (i.e.: number of dependent functions, MNA score). Variables analyzed under multiple binary logistic model were dichotomized as described before. Initially, all variables were included within the unadjusted models, using the hierarchical approach presented in Figure 1. Variables within each block were analyzed with regards their significance. Variables with p-value above 0.20 were progressively removed using the Backward-Wald method. A significance level of 5% was the criterion for a statistically significant effect. Odds ratios (ORs) were reported with 95% confidence intervals (95%CI). Statistical B-coefficient (regression coefficient) was considered to continuous variables included in the model. B-coefficient reported how much every increment of the independent variable would increase chances to have an HRQoL impact.