*2.2. Questionnaire*

The initial part of the questionnaire included questions that allowed for a reliable selection of study participants, dividing them into groups. The questions concerned the existing diseases, duration of the T1DM and type of treatment, sex, age, place of residence. The body weight and height (self-reported) results were used to calculate the body mass index (BMI), which reflected the general nutritional status of the patient. It was calculated as: weight in kg divided by height in meters squared. In children and adolescents under 18 years of age, it is interpreted according to national standards, and the limits of underweight, overweight

and obesity are defined as the 10th, 85th and 97th centiles, respectively [10]. For adults, the values established by the World Health Organization were applied: a person whose BMI is below 18.5 kg/m<sup>2</sup> is considered underweight, the normal value is 18.5–24.9 kg/m2, whereas in overweight and obese persons the values are 25.0–29.9 kg/m<sup>2</sup> and over 30.0 kg/m2, respectively [11]. The results of glycated hemoglobin (HbA1c) from the last 3 months (selfreported) were obtained in a laboratory at the request of the attending physician.

The next part of the survey included questions about lifestyle (sleep time, screen time, stress levels), physical activity and eating habits, including the Mediterranean Diet Adherence Screener (MEDAS), which consists of 14 questions about eating behaviors typical of a MD (Table 1). Each question could earn a point; the maximum number of points to be earned was 14. The responses were to refer to the last month preceding the completion of the questionnaire. Based on the total scores, participants were divided into three levels: low (score 0–5), medium (6–9 points) and high ( ≥10 points) MD adherence.

**Table 1.** Interpretation of Mediterranean Diet Adherence Screener.


Category: low (score 0–5), medium (6–9 points) and high (≥10 points) Mediterranean Diet adherence.

> The entire questionnaire consisted of questions that had appeared in our previously published study and other authors' work [12,13]. Questions in foreign languages were translated into Polish and assessed by a native speaker of the Polish language in order to exclude any bias in the translation. The translated questionnaire was tested on a small sample of respondents in order to avoid formal and substantive errors.

## *2.3. Statistical Analysis*

Statistical analysis of the results was performed using Statistica software (TIBCO Software Inc., Palo Alto, CA, USA). The Shapiro–Wilk test was applied to check the normal distribution of the variables. According to the test outcomes, Student's *t*-test (parametric variables), the Mann–Whitney U and Kruskal–Wallis ANOVA tests (non-parametric variables) were used. The Chi-square independence test evaluated the relationships between qualitative features. Before the survey, a required minimum sample size was estimated. It was useful for calculating the total participants of our study with a specified confidence interval (95%) and a maximum bias (10%). Values at *p* < 0.05 were considered statistically significant. The supplementary material contains additional characteristics of the most significant results divided according to variables (place of residence, age group).
