MENTAL WELL-BEING IN ADOLESCENTS WITH DIABETES DURING THE COVID-19 PANDEMIC AND IMPACT ON GLYCAEMIC CONTROL

Introduction : Psychological distress was experienced by millions of people affected by the coronavirus (COVID-19) pandemic, particularly adolescents with diabetes. Method : A cross-sectional study was conducted among adolescents with type 1 (T1DM) and type 2 diabetes mellitus (T2DM) aged 10-18 years during the nationwide lockdown from June-December 2020 in Malaysia to assess the impact of COVID-19 towards their mental health. Eighty-seven participants were recruited and interviewed regarding their lifestyle changes, followed by the administration of the Depression, Anxiety and Stress Scale (DASS-21). Results : Fifty-eight patients with T1DM (30 males, 51.7%) and 29 patients with T2DM (9 males, 31%) participated in the study. The incidence of depressive, anxiety and stress symptoms were detected in 34%, 41% and 26% respectively of diabetic adolescents with no significant difference between T1DM and T2DM patients. Male gender, T2DM, puberty and the presence of anxiety symptoms were associated with deterioration in glycaemic control post-lockdown. There was a deterioration in HbA1c among male and T2DM patients by 0.76% and 0.94% respectively (p-value 0.013, 0.004 respectively). An increment of HbA1c pre-and post-lockdown was observed in patients with anxiety symptoms (9.39 ± 0.49 vs 10.16 ± 0.54%, p-value 0.028). Patients who displayed stress symptoms showed improvement in their HbA1c (10.00 ± 0.57% vs 9.50 ± 0.063%, p-value 0.036). Conclusion : There is a high prevalence of psychological disturbance among diabetic adolescents during the COVID-19 pandemic, particularly the anxiety component related to poor glycaemic control. Follow-up psychological assessment and support must be given to these young diabetic patients.


Introduction
The COVID-19 pandemic in March 2020 has posed great challenges to every layer of society and healthcare systems worldwide. To limit the transmission of the disease, Malaysia implemented different stages of nationwide lockdowns during the pandemic.
Adolescents are particularly vulnerable to this dramatic disruption to their daily lives, as they are at a crucial stage of neurobehavioral maturation. School closures and home confinement restricted interactions with their peers. The continuous fear of being infected with the COVID-19 virus, and increased family stress pertinent to the pandemic, such as parental job and financial insecurity could further trigger negative psychological effects [1].
Diabetes adolescents have the additional burden of managing their diabetes and difficulties with medical access during the lockdown.
Thus, the primary aim of this study was to identify the impact of lockdowns towards mental health among diabetic adolescents. Secondly, to relate those changes during lockdown with the change of their glycaemic control.

Methods and descriptions of participants
This cross-sectional study was conducted in compliance with ethical principles outlined in the Declaration of Helsinki and the Malaysian Good Clinical Practice Guideline.
The study was approved by the Medical Research and Ethics Committee (MREC), Ministry of Health (NMRR-20-1678-55949).
A cohort of 87 patients with Type 1 Diabetes Mellitus (T1DM) and Type 2 Diabetes Mellitus (T2DM) were recruited between June till December 2020 during the nationwide lockdown. During the study period, the schools remained closed most of the time with high daily increasing positive cases of COVID-19 (refer to Figure 1). The inclusion criteria were T1DM and T2DM patients aged 10 to 18 years old under follow-up at the Paediatric Endocrine Clinic at Putrajaya Hospital with a duration of diabetes more than 6 months before the COVID-19 pandemic. Written consent was obtained from the parents and the patients. The HbA1c taken during the clinic visit between June till December 2020 was compared with the last HbA1c retrieved before the COVID-19 pandemic.
During clinic visits, they were interviewed regarding their demographic characteristics and lifestyle changes during the lockdown, followed by the administration of the Depression, Anxiety and Stress Scale (DASS-21).
Lifestyle changes before and during the lockdown were assessed by questionnaires, which were based on 4 domains: meal frequency, physical activity level, screen time and sleep adequacy. Malay version and local adaptation of the Physical Activity Questionnaire for Older Children (PAQ-C) and the Physical Activity Questionnaire for Adolescents (PAQ-A) were utilised [2]. "Sufficiently active" was defined as PAQ scores of more than 2.9 in boys and more than 2.7 in girls. Following the recommendation by Canadian 24-hour Movement Guidelines, recreational screen time should not be more than 2 hours per day for both children aged 5 years above and youth [3]. Duration of sleep was deemed adequate if it was 9 to 12 hours per day for children 10 to 12 years old, and 8 to 10 hours per day for adolescents 13 to 18 years old [4].
The depression, Anxiety and Stress Scale (DASS-21) is a self-reported instrument to assess the level of depression, anxiety and stress among the participants 5]. There are three subscales, namely depression, anxiety and stress with seven items on each subscale. Patients were required to score each item on the Likert scale ranging from 0 to 3 depending on the frequency of occurrence throughout the week (0: never, 1: sometimes, 2: often, 3: almost always). The sum score of each subscale was multiplied by the factor of 2. Therefore, the total score of each subscale ranged from 0 to 14 and the total score for the whole questionnaire ranged from 0 to 42. The cut-offs for depression (mild 10-13, moderate 14-20, severe 21-27, extremely severe ≥28), anxiety (mild 8-9, moderate 10-14, severe 15-19, extremely severe ≥20) and stress (mild 15-18, moderate 19-25, severe 26-33, extremely severe ≥34) were utilised. The validated Bahasa Malaysia version of DASS-21 was used in this study, as it showed good reliability and validity [6] in the multi-ethnic population in Malaysia.

Statistics
The data were analysed with Statistical Package for Social Sciences version 23.0 (IBM Corp., Armonk, NY, USA). The level of significance was set at p < 0.05. Numerical variables were summarised as mean and standard deviation and categorical variables were summarised as frequency and percentage. Change in HbA1c from before and during the lockdown across different groups was analysed using the One-way Repeated Measures ANOVA test. The level of significance was set at p < 0.05.  Among the T1DM patients, the incidence of depression was 32.8%, anxiety 46.6%, stress 31.0% and combination of subscales 38%. The incidence of anxiety (31.0%) and stress (17.2%) were lower in T2DM patients, while the depression rate was slightly higher (37.9%). In terms of severity, 20.6%, 31.1% and 14.9% of patients had moderate to extremely severe symptoms for the subscale of depression, anxiety and stress (refer to Figure 2).   Data were expressed as adjusted mean with standard deviation in parentheses. Continuous variables were analysed by One-way Repeated Measures ANOVA test using General Linear Model, after being adjusted by age (p-value < 0.001) and diabetes duration (p-value = 0.050).

Discussion
The COVID-19 pandemic and lockdowns implemented in many countries placed a considerable psychological toll on adolescents worldwide. The incidence of the significant score of depression (20.7%) in our cohort was slightly lower than the pooled prevalence rate of 25.2% in the meta-analysis done by Racine N. et al. [7]. The percentage of the significant score of depression was even lower in Italian youths with T1DM (16%) [8] and the researchers also found that the higher score of depression was associated with a lower time in range. However, we did not find any association between HbA1c with depressive symptoms.
Duan et al. [9] identified the association of addiction to social media to an increased level of depression: internet addiction (OR = 3.107, 95% CI: 1.252 to 7.708) and smartphone addiction (OR = 1.844, 95% CI: 1.29 to 2.811). The screen time in our patients increased more than two folds during the lockdown (before 2.30 ± 1.86 hours vs during 5.89 ± 2.84 hours).
Thirty-one per cent of our diabetic adolescents had a significant score of anxiety, which was higher than the pooled prevalence rate across 25 studies (20.5%) in the general population [7]. It was much higher than the Italian cohort that reported only 7% had a significant score of anxiety [8]. The possible explanation was that the psychological data in Italy were collected much earlier (March to April 2020) compared to our cohort. It was found that the prevalence of depressive and anxiety symptoms increased as the pandemic progressed [7]. In New South Wales, Australia, 45% of children with T1DM reported that they were worried about contracting severe COVID-19 illness as they had underlying comorbidity [10].
The incidence of a significant score of stress was 14.9% in our cohort. Social isolation stress may lead to motivation to consume a high fat diet ("comfort food") to alleviate depression-and anxiety-like behaviours in humans [11]. This could result in increased food intake and weight gain [12]. Similarly in our cohort, there was a significant weight gain during the lockdown (prepandemic mean weight 50.3kg vs 52.3kg, p <0.001). In addition, physical activity had reduced tremendously, mean of 249.9 minutes per week before to 175.6 minutes per week during the school closure (p = 0.038). Interestingly, our patients with stress symptoms had better glycaemic control during the lockdown. This could be due to the more stressed patients were also more worried about their diabetes control.
Traditionally, adolescent females have poorer glycaemic control and a higher incidence of microvascular complications compared to their male counterparts [13]. However, this genderdependent difference was reversed in our study with male adolescents having worse glycaemic control. This was possible due to poorer coping with indoor restrictions in adolescent males who are more physically active compared to females [14].
Similar to the findings by Eberle C and Stichling S [15], our T2DM patients had worsening glycaemic parameters during the lockdown (before 9.14% vs during 10.08%). These could be related to increased snack consumption, food cravings and physical inactivity during home confinement [16]. Socioeconomic difficulties experienced during the pandemic could also alter the choice of healthy nutrition [17]. An increment of 0.51% in HbA1c was observed in diabetic youths in South Carolina not receiving routine access to a diabetic care provider [18].
The strengths of the study were that the interviews were conducted by a single researcher to avoid bias. Standardized questionnaires were utilized for screening mental health issues and gauging lifestyle changes. Accurate clinical information was readily available from the hospital's Electronic Medical Report. All patients were managed by the same paediatric endocrinologists which are the authors. The limitation of the study was no baseline data available regarding the mental well-being of the participants before the COVID-19 pandemic for comparison. There was also no local data to compare the well-being of diabetic adolescents with non-diabetic adolescents during the pandemic. The long-term consequences of the COVID-19 pandemic towards mental well-being and glycemic control are not able to be addressed in this study due to the short study period.

Conclusion
Lifestyle changes and uncertainties during the COVID-19 pandemic have had a significant impact on the mental health of diabetic adolescents with a high prevalence of psychological disturbances reported. Male gender, type 2 diabetes mellitus and presence of anxiety symptoms were associated with deterioration of glycaemic control. This study highlights the need for the provision of adequate psychological services to this vulnerable group.