Study design
Cross-sectional evaluation carried out at the Diabetes Unit of University Hospital Virgen del Rocío (UHVR). UHVR is a tertiary care DM1 reference center, the largest complex of the Andalusian public healthcare system, serving a population of 554.981 inhabitants. At the time of this evaluation, 1,665 DM1 adolescent and adult patients were included in the UHVR Registry. Internal Review Committee at UHVR approved the study protocol. All patients signed informed consent prior to inclusion in the study.
Patients’ selection
Adult patients (ages 18 – 65 years) with DM1 [according to American Diabetes Association criteria (ADA 2014) [17] for at least one year, were randomly selected at UHVR using a computer-generated random list. Exclusion criteria included pregnancy or pregnancy planning; psychiatric or neurological disorders limiting their ability to complete the questionnaires and any non-diabetes related clinical condition that could affect the evaluation. Patients selected were informed about the aims of the study by their physician and if agreed to participate, they were contacted by phone (for up to three times) to schedule an appointment.
Quality of life evaluation
QoL was assessed by the self-administered Diabetes QoL (DQoL) questionnaire translated and validated into Spanish (Es-DQoL). The DQOL questionnaire showed high reliability as well as internal and external validity, both for the original version and the Spanish translation [18]. It contains 43 items that patients rank on 5-point Likert scale ranging from 1 to 5 (1 = never; 5 = all the time). Four subscales measure diabetes impact on daily life (17 items, range: 17–85), diabetes-related worries (4 items, range: 4–20), satisfaction (15 items, range: 15–75), and social worries (7 items, range: 7–35). The Es-DQoL score ranges between 43 (highest level of QoL) and 215 (lowest level of QoL). In order to facilitate data comprehension, the variable "Total EsDQoL score” and subscales scores were presented on a scale of 0-100, where the higher the score, the higher QoL, using the following formula: (Transformed score - minimum possible score)/(maximum possible score-minimum possible score) x 100, where “Transformed score” is calculated as maximum possible score - (real score-minimum possible score).
Clinical evaluation
Clinical and sociodemographic variables were collected at study entry; biochemical variables (HbA1c levels) were taken from medical records (most updated record within a 3-month period before study entry). Variables included age; gender; level of education; marital status; age at disease onset; disease duration; HbA1c levels; presence of chronic diabetes complications; ketoacidosis episodes and/or other diabetes-related hospital admission (excluding hospital admission at onset); severe hypoglycemic episodes, and total hypoglycemic and hyperglycemic episodes in the previous 15 days. Glycemic-related variables were collected from self-monitoring glucose data (patient electronic records or glucose diary). Glycemic instability, severe hypoglycaemia, unawareness hypoglycaemia, chronic diabetes complications, retinopathy, nephropathy and evaluation of diabetic peripheral neuropathy and finally, peripheral vascular complications have been defined previously [19-21].
Psychological evaluation
Psychological assessment was carried out at outpatient facilities and included a structured diagnostic interview by a psychologist included in the research team through Mini International Neuropsychiatric Interview (MINI) [22]. This planned interview aimed at detecting psychiatric disorders such as depressive and/or anxious symptoms. Self-reported questionnaires were used for grading severity of psychological symptomatology [Beck Depression Inventory II (BDI-II) (23), State-Trait Anxiety Inventory, Form Y (STAI-Y) (24,25)], detecting Fear of hypoglycemia (FH), [Fear of hypoglycaemia Scale (FH-15), cut-off score 28 (26)] and determining the social support received by patients [Medical Outcomes Study (MOS) Social Support Survey (27, 28)] according to standard procedures.
Statistical Analysis
Sample size was calculated in order to detect a minimum difference of 4 points at Total Es-DQoL score between male and female subgroups, a 95% confidence level, a statistical power of 90% and assuming a 25% drop-out rate. The required estimated sample size was 420 patients. Normally distributed variables are presented as mean and standard deviation (SD). Variables with skewed distribution (Es-DQoL and subscales) are presented as geometric means and interquartile range. In order to detect gender differences, qualitative variables were compared using chi-squared test and quantitative variables using t-Student test. Pearson´s correlation coefficient (rho) was used in univariate analyses for quantitative variables. A Stepwise multiple linear regression was applied to explore predictors for QoL in our sample of patients. Variables included in this analysis were significant in univariate analyses, demographic, clinical and disease-specific variables, as well as the scores on psychological tests. For the analysis of the data, the statistical package package IBM SPSS Statistics 22.0 for Windows and statistical significance was defined as ≤ p 0.05.