Are We Ready to Treat Our Diabetes Patients Using Social Media? Yes, We Are

Abstract

Aims:

The aim of the study was to evaluate Facebook group as possible communication tool to improve glucose control in adolescents and young people with type 1 diabetes (T1D).

Methods:

This retrospective and cross-sectional study included 728 T1D patients (age 11-25) on continuous subcutaneous insulin infusion (CSII) or multiple daily injections (MDI) treated at the center for insulin pump and glucose sensor in Skopje from January 2012 to December 2017. Data were collected through the electronic medical record system and cross-sectional analysis (telephone, social media (Facebook and Viber) or email). Patients were analyzed in two groups: (a) non-Internet group, with 398 patients treated using standard medical protocol with regular clinic visits; (b) Internet group, with 330 patients who, besides standard medical protocol, were active members of a national closed Facebook group on diabetes. Both the non-Internet and Internet groups had regular visits every 2-3 months. Patients from the Internet group were members of the closed Facebook group “Diabetes Macedonia” and had an opportunity to interact with questions, answers, and comments on diabetes care. An additional analysis was performed of the Internet group on combined use of Facebook and Viber. Average HbA1c levels were compared in both groups.

Results:

Each patient from the Facebook group had 1.5 ± 3.5 posts per day. Hba1c was significantly lower in patients from the Internet group (7.1 ± 3.2%; 54 ± 35 mmol/mol) compared to patients from the non-Internet group (7.6 ± 2.8%; 60 ± 31 mmol/mol).

Conclusions:

Social media like Facebook and Viber can be additional communication tool in adolescents and young people with T1D and can significantly lower HbA1c compared to patients without social media use. CSII patients are more likely to use both social media (Facebook and Viber) compared with MDI patients (Facebook only).

Successful management of type 1 diabetes (T1D) requires frequent glucose monitoring and insulin dose adjustment in response to various factors (carbohydrate intake, glucose values, physical activity, stress, illness), which can improve clinical outcomes in patients.¹ However, a high proportion of T1D patients report difficulties in their management associated with negative outcomes, which can impose a substantial burden that negatively affects patients and family quality of life.²

One potential solution in improving T1D management is the use of technology, providing additional opportunities to support management and maintain and improve communication and engagement with health care services. Patients use the Internet to find specific health information, share their information, provide and receive support, and interact with other patients.^3-5 Health-related websites provide one-way communication where users search for specific information.

Social media is a two-way communication tool, which enables social support and interaction in the online community. It provides opportunities for users to generate, share, receive, and comment on social content among multiuser community.⁶ The social media platforms can overlap among the various services that allow different communication channels like social networking (Facebook), microblogging (Twitter), instant messaging (WhatsApp), video sharing (YouTube), photo sharing (Instagram), telephony (Skype), and file sharing (Dropbox). The social media platforms and discussion forums are very popular among young people,⁷ providing unique opportunities for online diabetes education, intervention, and support.^8-10

The positive impact of social media use in health care identifies different effects and outcomes: it fosters patients’ autonomy by complementing the information provided by health professionals (HP);¹¹ facilitates patient to HP dialogue; provides psychosocial support;¹² enhances patient’s empowerment;¹³ collects data on patient experiences and opinions; promotes health and health education; improves health and behavior change;¹⁴ and reduces illness stigma. Patients are willing to share highly personal information on social media, including thoughts, feelings, and behaviors that they typically would not disclose because of stigma or embarrassment.^15,16

However, social media use by patients provides not just beneficial effects. It also incurs the increased risk of disseminating misinformation and also has some limitations: lack of reliability; quality concerns; lack of confidentiality and privacy; risks of disclosing personal information online; and harmful or incorrect advice. There is a risk of “unreliable post advice” on certain health condition, which can lead to unwanted health problems. Social media can be viewed as manipulative and often perceived as business tool to sell goods and services.

Facebook, with over 2.19 billion active monthly users worldwide,¹⁷ is an important source of information, support, and engagement for patients with chronic disease. Facebook groups can be used as a supportive tool in disease management for patients and families,¹⁸ which is not necessarily available through formal professional consultation.¹⁹ A recent study evaluates the use of Facebook as a communication tool between patients and families where provided information led to better T1D management in children.²⁰ The communication between patients and HP facilitated clinical personalization and allowed active patient participation in the decision-making process with improved glucose control in T1D patients using insulin pump.²¹

Our diabetes center uses social media (Facebook and Viber) in communication with our patients. Our closed Facebook group on diabetes (Diabetes Macedonia), formed in 2008, has 1256 members, and the Viber group “Team Type 1” has 92 members. In our hospital, health care personnel routinely ask all patients with T1D to join the groups (Facebook and/or Viber) in the formal clinic consultation, where access is granted to prevent fake users from joining the groups.

The aim of the study was to evaluate results of Facebook group as possible communication tool to improve glucose control in T1D adolescents and young people and to initiate a discussion of possible benefits of social media use in diabetes care.

Methods

A single-center cross-sectional and retrospective analysis was conducted at the Center for Insulin Pump and Glucose Sensor, University Clinic of Endocrinology, Diabetes and Metabolic Disorders in Skopje. The study enrolled 728 T1D patients (age 11-25 years) treated at our center from January 2012 to December 2017. Data were collected through the electronic medical record system, cross-sectional analysis (telephone, social media [Facebook and Viber]), or email.

Patient selection criteria were T1D on multiple daily injections (MDI) or continuous subcutaneous insulin infusion (CSII) and age between 11 and 25 years.

Patients were excluded from the study if they had fewer than three visits per year and T1D duration less than 6 months. According to our standards of care, T1D patients are evaluated four times per year by a diabetes multidisciplinary team. Informed consent and assent form (patients <18 years old) were obtained from the patients and their caregivers. The research plan has been approved by the local Ethics Committee.

Patients were analyzed in two groups:

• Non-Internet group: 398 patients were treated using standard medical protocol with regular clinic visits

• Internet group: 330 patients, besides standard medical protocol were active members of national closed Facebook group on diabetes

Both the non-Internet and Internet groups had regular visits every 2-3 months with standard medical protocol. Patients from the Internet group were members of the closed Facebook group “Diabetes Macedonia” and had an opportunity to interact, asking and answering questions on diabetes care. Two physicians and three diabetes nurses administered the group on a voluntary basis. An active member on Facebook was considered if the patient had at least three posts per month in a period of one year.

Demographic and clinical data (sex, age, weight, diabetes duration, and HbA1c level) were collected through the electronic medical record system. The cross-sectional analysis was conducted from January 2017 to December 2017 by telephone, email, or social media (Facebook and Viber) to collect actual data including total daily insulin (TDI), diabetic ketoacidosis (DKA) events, and severe hypoglycemia. Social media use and members’ posts on Facebook and Viber were also analyzed in the last five years (January 2012 to December 2017). The posts and discussion were grouped in five segments: carbohydrate counting, low and high blood sugar treatment, CSII troubleshooting, new treatment options, and other issues. Additional analysis was performed on patients who used the Facebook group “Diabetes Macedonia” and the Viber group “Team Type 1.”

Statistical analysis was performed with SAS version 12.5 for Windows (SAS Institute, Cary, NC). Quantitative variables are expressed as mean ± standard deviation. Frequency distributions and appropriate summary statistics for central tendency and variability were used to describe possible differences between the two groups. The HbA1c variation between groups was analyzed using the paired Student t-test. A P value < .05 was considered to denote significance.

Results

Among 1028 T1D patients, 728 patients met criteria for analysis. Patients were analyzed into 2 groups: 398 patients in the non-Internet group (standard medical protocol) and 330 patients in the Internet group (standard medical protocol plus active member on Facebook group). Demographic and clinical characteristics of this population are described in Table 1. We found a significant difference in CSII users in the Internet group compared with the non-Internet group (39.7% vs 10.6%).

Table 1.

Demographic Characteristics of the Patients.

Group	Non-Internet group	Internet group
Number	398	330
Age, years	15.2 ± 2.9	16.4 ± 1.9
Male	173	141
Female	225	189
Diabetes duration, years	5.9 ± 2.4	6.3 ± 2.1
CSII use, %*	10.6	39.7
BMI, kg/m2	21.7 ± 3.4	22.4 ± 2.9

Values are expressed as mean ± standard deviation, unless otherwise noted.

^*Significant difference (P < .05).

Patients from the Internet group had a significantly lower HbA1c compared to patients from the non-Internet group, one year after joining the closed Facebook group (as shown in Table 2). We did not find significant difference in DKA events, severe hypoglycemia, and TDI between both groups.

Table 2.

Clinical Characteristics of the Patients.

Group	Non-Internet group	Internet group	P
HbA1c, %	7.6 ± 2.8	7.1 ± 3.2	<.05
HbA1c, mmol/mol	60 ± 31	54 ± 35	<.05
Diabetic ketoacidosis, per patients/year	0.04 ± 0.3	0.02 ± 0.1	NS
Severe hypoglycemia, per patients/year	0.01 ± 0.2	0.01 ± 0.2	NS
TDI, U/kg/d	0.84 ± 2.6	0.76 ± 1.5	NS

Values are expressed as mean ± standard deviation, unless otherwise noted.

Patients were active on the Facebook group with 1.5 ± 3.5 posts per patient per day. The most discussed topics were carbohydrate counting (32%), low and high blood sugar treatment (26%), CSII troubleshooting (21%), new treatment options (16%), and other issues (5%). Health providers (physician, diabetes nurse) responded and clarified 26% of patients’ questions and comments. Most of the patients (64%) shared their last HbA1c in a group (Figure 1), which was used as a motivational and supportive tool for other patients (each HbA1c post had on average 8.9 replies).

Figure 1.

Social media support among T1D patients. (A) Facebook: motivation for better control; (B) Viber communication: glucose sensor problem solving.

An additional analysis was performed on the Internet group on combined use of Facebook and Viber (as shown in Table 3). Patients using Facebook and Viber were mostly CSII users and had significantly lower HbA1c (6.7 ± 2.1% vs 7.3 ± 1.9%) and were more active in communication on social media compared with patients using Facebook only (1.8 ± 3.6 vs 1.1 ± 2.3 posts per day).

Table 3.

Combined Use of Facebook and Viber in Internet Group

	Facebook only	Facebook and Viber	P
N	246	84	<.05
Posts per patient per day	1.1 ± 2.3	1.8 ± 3.6	<.05
CSII users, %	29.8	70.2	<.05
HbA1c, %	7.3 ± 1.9	6.7 ± 2.1	<.05
HbA1c, mmol/mol	56 ± 21	50 ± 23	<.05

Values are expressed as mean ± standard deviation, unless otherwise noted.

Discussion

In this study, we showed that adolescents and young people with T1D who use social media in everyday practice achieved better control compared to patients without social media use.

One of the most important findings in the present study is the significant difference in mean HbA1c of 0.5% between patients who use and patient who don’t use social media in diabetes management. T1D patients express a high interest in using social media,^22,23 but clinical evidence in improving diabetes care is mixed. A recent publication reports significant increase in blood glucose monitoring per day without significant improvement in glucose control.²⁴ Another study on online chat group shows increased life satisfaction without improvement in glucose control.²⁵

We found a significantly lower HbA1c level in patients using Facebook and Viber, compared to patients using Facebook only. Patients using both forms of social media (Facebook and Viber) were more active in communication, with almost 50% more posts than patients using Facebook only. CSII patients were more likely to use both Facebook and Viber communication in diabetes management. Our findings are similar to recent studies, where patients preferred to use social media through moderated groups,²⁶ rather than with personal social networks.^26,27

A recent survey reported that 50% of T1D patients were engaged in social media communication once a day.²⁸ Patients in our study were active on social media with an average of 1.5 posts per day. In terms of active posts on social media, we found focused discussion on carbohydrate counting, low and high blood sugar treatment, and glucose monitoring, where patients look for advice. In our study, most of the patients shared their recent HbA1c on the Facebook group “Diabetes Macedonia” by themselves, which had a positive impact on patient empowerment and motivation. A similar study reported that the most active posts were on sharing personal experience, supporting patients’ care, and providing spiritual support.²⁹

Concerning safety issues, we found a low incidence of DKA and hypoglycemic events in both groups, which can be compared with similar results published by our center.^21,30

Our study has several limitations, including a retrospective design and single center study. However, the strengths of our study are large sample size and its real-life conditions.

Before implementing social media as a communication tool among patients and families, health providers need to address whether they are ready and capable to handle this additional information about their patients. An integrated system between social media and the health care system can be overwhelming with 24/7 online support; even more frightening are the possible risks that could occur from being one of the first to implement this approach. There are at least two critical points that physicians cannot normally offer: the perspective from the patients’ point of view and an almost unlimited amount of time to listen and share experience.

We believe that in today’s challenging health care environment of limited budgets and resources with a desire to provide better diabetes care, new methods of patient interaction using social media can be beneficial.

Conclusion

Social media like Facebook and Viber can be additional communication tool between adolescents and young people with T1D and HP and can lead to improved glucose control. CSII patients more often use social media in communication and have significantly lower HbA1c compared to patients without social media use. Further randomized studies with larger population and multiple sites are required to confirm the potential benefits of social media use in diabetes management.