Characteristics of the participants

A total of 1011 responses to the online questionnaire were collected: 485 from those responsible for children and adolescents with diabetes (diabetes group) and 526 from those responsible for children and adolescents without diabetes (non-diabetes group). After excluding participants who did not agree with the informed consent, did not meet the inclusion criteria or completed less than 75% of the questionnaire, 381 participants in the diabetes group and 383 in the non-diabetes group were included in the analyses (see supplementary Fig. 1).

Overall, the participants included in the study had a mean age of 39.9 ± 8.5 years; 95.2% were female, 78.3% were white and 47.8% had medium–low family income. Regarding the relationship with the child, 89.1% were mothers, either biological or adoptive. The diabetes and non-diabetes groups did not differ in relation to the participants’ age, sex, and relationship with the child. There was a higher prevalence of non-white and low-income participants in the diabetes group (31.2% vs. 12.3%, P < 0.001 and 54.1% vs. 41.5%, P = 0.001, respectively) (see Table 1). Representatives from all regions of Brazil were included, although a high proportion of participants in both groups were from the South-Southeast regions (79.0% in diabetes group and 97.4% in non-diabetes group).

Table 1 Demographics and clinical characteristics of study participants.

Children and adolescents in the diabetes group were older (11.8 ± 4.3 vs. 8.1 ± 4.7 years, P < 0.001); 28.5% of the participants in the non-diabetes group reported the presence of a chronic illness in the child. Among these participants, 83 had respiratory tract diseases (asthma, bronchitis, cystic fibrosis or rhinitis requiring continuous treatment), 3 had congenital heart disease, 7 had neuro-psychiatric diseases (autism spectrum disorder, attention deficit and hyperactivity or anxiety in need of pharmacological treatment), 1 had epilepsy, 1 polycystic kidney disease, 1 lactose intolerance, 1 deafness and 1 cerebral palsy. Of these, 15.1% used medication daily.

Support, relationships, and consequences of social distancing

Considering the social-demographic differences between the two groups, the OR for the characteristics evaluated in the psychosocial profile were adjusted for age of the child, race/ethnicity, income, and region of origin. Participants in the diabetes group more frequently reported unsatisfactory relationships (18.0% vs. 9.4%, adjusted OR 1.90; 95% CI, 1.16 to 3.12) and regarding family atmosphere, the diabetes group more frequently reported the atmosphere as not-welcoming (8.0% vs. 1.3%, adjusted OR 5.85; 95% CI, 2.09 to 16.43). In addition, a greater purchasing difficulty (38.7% vs. 10.8%, adjusted OR 4.65; 95% CI, 2.93 to 7.38) and a greater difficulty in getting medical assistance when necessary during the pandemic (44.2% vs. 18.5%, adjusted OR 3.36; 95% CI, 2.27 to 4.97) were reported by the diabetes group compared to non-diabetes group. Aspects related to the social distancing, the child being at home full time and reduced family income during the pandemic did not differ between groups (see Table 2). When the evaluation was carried out comparing only to healthy controls, there was no difference between the groups regarding the maintenance of unsatisfactory relationships, but the diabetes group more frequently reported the family atmosphere as not-welcoming (8.0% vs. 0.4%, adjusted OR 17.39; 95% CI, 2.24 to 134.99). There were few differences in the other parameters (see supplementary Table 2).

Table 2 Support, relationships, and consequences of social distancing in study participants.

Pandemic-related emotional burden

The emotional burden evaluation during the COVID-19 pandemic was performed in four domains. Participants in the diabetes group most often expressed personal concern (84.4% vs. 78.3%, P = 0.041), child-related concern (92.6% vs 86.0%, P = 0.005), personal emotional burden (78.2% vs 65.3%, P < 0.001) and child-related emotional burden (75.2% vs. 57.1%, P < 0.001) when compared to the non-diabetes group (see Fig. 1).

Figure 1
figure1

Assessment of pandemic-related emotional burden between the diabetes group and the non-diabetes group.

The evaluation of pandemic-related emotional burden between the diabetes group and the non-diabetes group was performed in four domains using the Likert scale of 5 points of agreement. The domains were evaluated through the statements: (1) personal concern—“I often feel worried and afraid of being infected with the coronavirus”; (2) concern related to the child—“I often feel worried and afraid that my child may be infected with the coronavirus”; (3) personal emotional burden: “I often feel tired and exhausted due to the change in routine since the social distancing started due to the COVID-19 pandemic”; (4) child-related emotional burden: “I often feel tired and exhausted from the responsibility to protect my child during the COVID-19 pandemic”. It was considered “totally agree” and “agree” as an affirmative answer.The graphs show the percentages of people with affirmative answers to the proposed statements in each group. *P = 0.041; **P = 0.005; ***P < 0.001.

When adjusted by the age of the child, race/ethnicity, region of origin and income, the diabetes group maintained a greater likelihood of personal concern (OR 1.62; 95% CI, 1.01 to 2.58), a greater likelihood of concern for the child (OR 2.01; 95% CI, 1.11 to 3.64), a greater likelihood of personal emotional burden (OR 1.67; 95% CI, 1.10 to 2.53) and a greater likelihood of burden related to childcare (OR 2.28; 95% CI, 1.54 to 3.38).

A sensitivity analysis, adjusted by the same factors and excluding caregivers of youth with chronic disease in the non-diabetes group, was performed. The diabetes group maintained a greater likelihood of personal concern (OR 1.64; 95% CI, 1.02 to 2.64), a greater likelihood of concern for the child (OR 2.49; 95% CI, 1.35 to 4.59), a greater likelihood of personal emotional burden (OR 2.15; 95% CI, 1.41 to 3.27) and a greater likelihood of burden related to childcare (OR 2.62; 95% CI, 1.74 to 3.92) (see Supplementary Fig. 2).

Mental health during social distancing

The presence of a positive screening for mental health disorders was found in 50.3% of the participants in the non-diabetes group and in 69.0% of the participants in the diabetes group (P < 0.001 for the difference between groups). When adjusted for the age of the child, race/ethnicity, income and region of origin, the diabetes group was found to have a significantly higher likelihood of positive screening for mental health disorders when compared to the non-diabetes group (OR 2.43; 95% CI, 1.70 to 3.47).

A sensitivity analysis, adjusted by the same factors and excluding caregivers of youth with chronic disease in the non-diabetes group, was performed. The diabetes group maintained a significantly higher likelihood of positive screening for mental health disorders when compared to the healthy control group (OR 2.68; 95% CI, 1.82 to 3.96) (see Supplementary Table 2).

Diabetes-specific emotional burden related to diabetes care

In the diabetes group, the diabetes-specific emotional burden related to the care of a child/adolescent with diabetes was evaluated in five domains: 40.6% reported discontent in care sharing, 36.0% reported discontent in support, 41.8% reported discontent in appreciation, 48.3% reported exhaustion and 75.7% reported guilt problems (see Fig. 2).

Figure 2
figure2

Assessment of diabetes-specific emotional burden related to diabetes care in caregivers of children and adolescents with a previous diagnosis of type 1 diabetes during the COVID-19 pandemic.

The evaluation of diabetes-specific emotional burden related to diabetes care was performed using the Likert scale of 5 points of agreement. The domains were evaluated through the statements: (1) sharing of care: “I feel frustrated because I am the only one responsible for helping my child use medications and manage glycemic control.”; (2) support: “I feel frustrated with the lack of understanding and support I get from friends and family in relation to taking care of someone with diabetes.”; (3) appreciation: “I feel underestimated for all the effort I put into helping my child to take care of diabetes.”; (4) exhaustion: “I feel that my child’s diabetes is consuming a lot of my physical and mental energy every day.”; (5) guilt: “I feel guilty if my child’s diabetes is not well controlled.” It was considered that it is a problem if the answer was “totally agree” and “agree”, and it is not a problem if the answer was “strongly disagree” or “disagree”.

An exploratory analysis of subgroups was performed to identify the likelihood of emotional burden related to care of a child/adolescent with diabetes in certain interest groups. The presence of unsatisfactory relationships and family environment that is not welcoming were predictors of having a greater likelihood of reporting emotional burden related to care sharing, support, appreciation, and exhaustion in diabetes care, while the presence of a positive screening for mental health disorders was a predictor for burden in all areas of diabetes care evaluated (see Table 3).

Table 3 Subgroup analysis to assess predictors of risk of diabetes-specific emotional burden during the COVID-19 pandemic among guardians of children and adolescents with type 1 diabetes.

The impact of the pandemic on different age groups

For the assessment of psychosocial profile and consequences of social distancing on different age groups, the results were adjusted for age of the child, race/ethnicity, income, and region of origin. When considering the different age range, the group of caregivers of youth aged 6 to 12 years with type 1 diabetes most often had unsatisfactory relationships (OR 7.86; 95% CI, 3.59 to 17.22) and had a not-welcoming family atmosphere (OR 5.98; 95% CI, 1.27 to 28.06) in comparison to the non-diabetes group. There was no difference in these aspects in the other age groups. In addition, in all age ranges, caregivers of youth with diabetes had a greater likelihood of having difficulty purchasing during the pandemic. In those aged between 6 and 12 years and those older than 12 years there was a greater likelihood of presenting greater difficulty in getting medical assistance when necessary (OR 3.84; 95% CI, 2.12 to 6.94 and OR 3.69; 95% CI, 1.76 to 7.71, respectively) (see supplementary Table 3).

The emotional burden evaluation on different age ranges was also performed in four domains and adjusted for confounders. The two groups were similar in all domains when the dependent youth was younger than 6 years. On the other hand, there was an increase in likelihood to present child-related concerns (OR 3.19; 95% CI, 1.23 to 8.27 and OR 3.26; 95% CI, 1.22 to 8.71), personal emotional burden (OR 2.05; 95% CI, 1.17 to 3.58 and OR 2.61; 95% CI, 1.34 to 5.10) and child-related emotional burden (OR 2.79; 95% CI, 1.62 to 4.82 and OR 2.56; 95% CI, 1.35 to 4.82) among caregivers of youth with type 1 diabetes aged 6 to 12 years and older than 12 years, respectively (see supplementary Fig. 3).

Regarding mental health disorders, the diabetes group showed greater likelihood of presenting disorders in those aged under 6 years old (OR 2.92; 95% CI, 1.08 to 7.94) and in those aged between 6 and 12 years old (OR 2.89; 95% CI, 1.73 to 4.84). There was no difference between groups in caregivers of adolescents older than 12 years (see supplementary Table 3).



Source link