Effect of Screen time on Glycaemic control of Type 2 Diabetes patients during COVID-19 Outbreak: A Survey based Study

Introduction

Screen-time has been linked to adiposity and insulin resistance [1]. Evidence suggests CVD risk factors are present to a significantly higher degree in the subset of overweight/obese individuals that are also insulin resistant [2]. A multivariate analysis, after adjusting for smoking, age, exercise levels per day and other lifestyle factors in a prospective cohort study of 6 years, showed that there was 23% (95% confidence interval [CI], 17%-30%) increase in obesity with each 2-h/day increment in TV watching and a 14% increase in risk of developing diabetes [3]. There has been elevated all-cause and cardiovascular disease mortality risk, even after physical activity adjustment, with increased sitting time/sedentary behaviours [4]. The outbreak of COVID-19 has led to reduced physical activity because of higher chances of getting infection on venturing outside home [5]. It is presumed that prolonged staying at home due to this COVID-19 will increase sedentary habits like increased time spent on mobile phones and watching TV [6]. The WHO has set a target time to be spent for exercise in adults and the elderly, which is estimated to be 150 min/week of moderate physical activity [7].

Taking this idea into consideration, a cross-sectional survey-based analysis was conducted to see the effect of screen time spent by a T2DM patient during this COVID-19 pandemic on the glycemic parameters.

Materials and methods

This was a cross sectional study in which data were collected through two rounds of surveys (with answer options) sent to regular outdoor T2DM patients visiting the authors’ clinics from May 15 to June 26, 2020, via patients’ pre-registered email IDs. The local ethical body to which the authors are attached (Remedy Clinic Study Group) approved the study.

There is no ready set of questionnaire available on association between social media and glycaemia during this lockdown period and hence, the authors used a new set of questionnaires in English language, the idea of which were taken from the Diabetes Self-Management Questionnaire (DSMQ) [8], the Maastricht Study [9] and, also from a Saudi Arabian study [10].The authors validated their questionnaires by three independent expert physicians.

Patients’ last visits’ electronic health record was available in the author’s clinic database to look for the patients who were previously compliant with follow-up and medicines and those patients who did mention their previous social media time spent use, as a routine questionnaire in clinic, less than 1 hour were included. The patients matching baseline inclusion criteria were contacted telephonically before the emails were sent. Since many patients were not able to come to regular check-ups during the COVID-19 outbreak, but many did their routine tests, questionnaires were shared to collect the data.

The questions were provided with easy and flexible answer options (using categorical variables) for the ease of the patients who might not be able to write down the exact values. After contacting the selected group of patients telephonically, emails were sent for written informed consent form along with the survey questionnaires (figure 1). The questionnaires were distributed via email and patients were asked to revert within 10 days.

• Download figure
• Open in new tab

Figure 1

Survey questionnaires with answers

Inclusion criteria: a) T2DM patients; b) BMI 20 – 28 Kg/m², c) age 19 – 60 years; d) estimated glomerular filtration rate (eGFR) as per CKD-EPI of >60 ml/min/1.73m²; d) duration of diabetes 5 – 7 years and, f) No injectable in existing treatment regimen. Exclusion criteria: a) if there was any history of hospitalization in last three months and, b) if they did not give informed consent to share data.

A total of 344 (first round of the survey – 158 participants; the second round of the survey – 186 participants) patients participated. The first round of the survey was carried out from May 15 to May 29, and the second-round survey was carried out from June 15 to June 26, 2020. In the first round of the survey, 50 patients did not mention one of their glycemic parameters like HbA1c, FBS, or PPBS, and in the second round, 65 patients did not report any one of the above parameters. Hence their data were excluded from the final analysis. Finally, data of 108 patients from round one of the survey and 121 patients’ data from the second round of the survey were available for analysis. The questions used in the survey with their answer options are given in the appendix.

Sample size calculation

Due to paucity of published data related to the present study on the effect of screen time on glycaemic status in the intended study population from Eastern India, the sample size could not be computed by conventional sample size formula. Hence the sample size is based on the Thumb’s rules. When no baseline information for computation of sample size is available, Thumb rule states a minimum of 120 subjects for a cross-sectional survey. This sample size is bare minimum to carry out statistical data analysis after completion of the study [11].

Statistical analysis

SPSS software version V21 ® was used to perform Binary logistic regression for calculating the Odds ratio (OR) of the categorical variables. The outcomes, looked for in the study, were poor control of glycemic parameters like HbA1c (defined by >7%), FBS (defined by >150 mg/dL) and, PPBS (defined by >200 mg/dL) and the exposure variables were screen time spent by the person per day for all the three glycemic parameters, doctor’s visit and daily exercise for HbA1coutcome. Descriptive analysis was performed, and the data were expressed as numbers and percentages. Comparison between the proportion of the study parameters before and during the lockdown has been carried out by extended McNemar test.

Results

The baseline characteristics of the total analysed cohort is summarised in table 1.

The changes of the patients between pre-lockdown period and lockdown period is calculated by extended McNemar test which shows the glycaemic indices during lockdown period to be significantly higher than previous state and are depicted in table 2.

A total of 173 patients had a screen time of less than 2 hours/day in our study sample. Among the 173 patients, 73 (42.2%) had achieved HbA1c less than 7%, whereas the remaining 57.8% had HbA1c more than 7%. On the other hand, 56 patients had a screen time of more than 2 hours, out of which 44 (72.73%) had HbA1c more than 7%. Among the 173 patients, only 89 (51.44%) had an FBS value of more than 150 mg/dL as compared to 46 (82.12%) with a screen time of more than 2 hours. Out of these 173 patients, only 43 (24.86%) had a PPBS value of more than 200 mg/dL as compared to 41 (73.21%) with a screen time of more than 2 hours. By conducting binary logistic regression, we found that the odds of having a poor glycaemic control as per HbA1c, FBS and PPBS is 2.67 times higher (95%CI: 1.91-6.95), 4.34 times higher (95%CI: 1.52-4.76) and, 8.26 times higher (95%CI: 4.26–11.83) in the cohort with a screen time of more than 2 hours as compared to the cohort with a screen time of less than 2 hours, respectively (table 3).

The OR calculated for HbA1c (outcome variable) taking exposure variables of daily exercise time >30 minutes/day as standard cut-off and regular doctors’ visits, were 3.91 (95%CI: 1.22-2.76) and 2.73 (95%CI: 1.12-2.98), respectively. This means that the cohort having spent less than 30 minutes/day exercising had 3.91 times higher odds of having poor HbA1c control, and those not going for regular check-up also had a 2.73 higher odds of not achieving target HbA1c of < 7% (table 4).

Discussion

Covid-19 has considerably changed our lifestyle and has reduced our physical activity [5], leading to increased sedentary habits [6]. This study tried to shed some light on an entirely new dimension of COVID-19 induced reduced physical activity, increased screen time, and their effect on HbA1c and other glycemic parameters. We deliberately took a duration of diabetes of 5 – 7 years since that will help us get a population where there will be a chance of optimizing oral anti-diabetic drugs (OADs), and hence the inclusion criteria of no injectable can be justified. The age criteria of 19 – 60 years relate to an active, healthy cohort who can carry out exercise or lifestyle changes as per physician advice. It has been found that only 54.6% of Indian T2DM patients comply with dietary advice, and only 37.2% follow the advice of exercise [13]. A meta-analysis of controlled clinical trials have shown a significant reduction of HbA1c in the arm doing exercise V/s. the control group (7.65% V/s. 8.31%; weighted mean difference, –0.66%; P<.001) [14]. Another meta-analysis found no significant difference between aerobic exercise and resistance exercise after sensitivity analysis (p = 0.14) [15]. Interleukin 6 (IL-6) and C-Reactive Protein (CRP) have been used as surrogate systemic inflammatory markers associated with T2DM [16]. A study with 60 obese women undergoing lifestyle modifications by physical activity and weight loss, showed a marked reduction in inflammatory markers like IL-6 (p = 0.009), interleukin 18 (IL-18) (p = 0.02) and, CRP (p = 0.008) [17].

In this era of mobile devices, text messages and telephonic communications play a significant role, and studies have demonstrated a similar efficacy of short message service (SMS) by cellular phone and telephonic follow-up by a nurse in reducing HbA1c [18]. A prospective, parallel-group, randomized controlled trial conducted in Asian Indian men with impaired glucose tolerance showed a significant decrease in developing overt T2DM in the intervention group, which received lifestyle advice at regular intervals via SMS (n = 271) than the control arm (n = 266), p-value 0.015 [19]. Similarly, a web-based survey done on both type 1 (n = 549) and type 2 diabetes (n = 210) found a significantly high self-care behavior score with diabetes app users after adjustment of confounding factors, for type 1 the increase was 1.08 (95%CI: 0.46-1.7) units and for T2DM it increased by 1.18 (95%CI: 0.26-2.09) units [20]. Not only positive but also negative impacts are seen in many systematic reviews that are done, and we often get mixed responses [21].

This study found an increase in HbA1c, FBS and, PPBS with increased time spent on social media, reduced physical activity, and irregular physician follow-up.

Study limitations

There are many limitations in this study:

1. As a survey-based study, we had to depend solely on the patient’s data and could not verify it.

2. BMI could not be adjusted since it was not included in the answer sheet and, the electronic health record had the last visit BMI which might have changed during the survey time due to a sedentary lifestyle during the lockdown period.

3. We had to use a range for all the questions (categorical variable) since patients find it easy to answer, hence we could not get the exact value of all parameters and thus could not do a multiple linear regression analysis to look for adjustments or other variables that could have affected the outcome.

Conclusion

Since the outbreak of this pandemic, our life changed to a considerable extent. We need to tackle this problem with new ways as per suggestions by the international bodies, and T2DM patients should carry on their daily activity at home or in nearby areas taking proper recommended precautions. Rare papers are there on screen time spent and glycaemic outcomes in T2DM patients during this lockdown period. There seems to be an increased risk of uncontrolled glycaemia if the patient spends more time on social media with reduced physical activity, without any productive events, and these sort of sedentary changes are enhanced in these types of outbreaks. Social media can be used in many useful ways to communicate with the patients and provide awareness, but the non-judicious use might become a problem.