**5. Use of information communication technologies (ICT) for patients with chronic diseases**

Patients with chronic diseases need to monitor and record several biometric health parameters. For this purpose, in current health care system, they mainly note observations on a paper, although most devices in use enable storage of such biometric measurements. While there are issues with reading noted values from the papers or even with losing the papers, there are issues with transferring the stored data on the devices to general practitioners (GPs) or specialists. Moreover, the doctors would appreciate to monitor more parameters that are relevant for a holistic care.

On the market, there are devices that can automatically measure and transfer the measure‐ ments to the smartphones and dedicated servers: Fitbit wristlets measure user activity and sleep periods [31]; BitBite measures user nutrition habits [32]; iHealth [33], VitaDock [34], VPD [35], and Abbott [36] products measure temperature, blood pressure, pulse, blood oxygen saturation, and glucose level. The current trend is geared toward wearables and gadgets that help diagnose very specific diseases, such as peripheral neuropathy [37] and retinopathy [38], toward gadgets that can measure several parameters [39] and toward integration of function‐ alities from gadgets to smartphone applications, such as in Google Fit [40] and in Moves [41]. These applications perform comparably well as standalone devices [42].

Furthermore, there are numerous smartphone applications, working with manual data input or data from previously mentioned devices, that assist users in managing their health [43, 44] or diabetes in particular [45, 46]. Such applications do not only display the status of health parameters, but also provide personalized recommendations based on the input data. They mostly encourage users to change their behavior [47].

However, wearable devices and smartphone applications are only facilitators and not drivers of patient empowerment. The design of engagement strategies is more important for successful use and potential health benefits of these devices than the features of technology [48].

Several pilots have been conducted, suggesting positive effects on health and diabetes care, and a need for 24/7 support [49–51]. However, in use, there are mainly only solutions that enable patients to informatively monitor their health status. Solutions that would be used as a part of the general health care service are in the stage of pilots and are only very rarely deployed as part of the standard practice.

## **6. Why mobile health?**

Mobile health is defined as a use of mobile communication devices, which include mobile phones, patient monitoring devices, tablets, personal digital assistants, and other wireless devices for health services, and information [52]. Currently, growing use of those devices can be seen in practically every part of the world. Number of mobile phone owners in the United States has reached 92% of a population and even number of smartphone owners has grown to 68%, while 45% of people own a tablet computer. Desktop or laptop computers were bought by 73% of Americans [53]. Surprisingly, similar rise in mobile technology use is also recorded among people in developing countries, where average share of people with mobile phones is around 83% [54]. With a vast majority of world population having an access to some type of mobile technology, this can certainly become a widely used to deliver deferent health care solutions to people. Role of mHealth is very broad and includes education and rising aware‐ ness, remote data collection, remote monitoring, communication with health care workers, support with diagnostic and treatment, and tracking diseases and epidemic outbreaks [55]. Most of these tasks are already performed by various mHealth applications for diabetes selfmanagement.

#### **7. Features of diabetes apps**

There are different classifications of diabetes app features and in this section we present some of them.

Review of accessible diabetes applications has shown that they mostly focus on blood glucose monitoring, medications, physical exercise, and diet management, while they also include other features such as education, communication, weight or BMI and blood pressure tracking, integration with public health records, decision support systems, and social networking. Blood glucose monitoring is available in all reviewed applications, while other features are more rarely present. Educational tools are brought to use in just 18% of applications and only around 30% of applications offer means to monitor weight, blood pressure, and physical exercise [56]. Still, among all medical conditions, diabetes with weight control represents the most addressed medical issue by mHealth applications in mHealth research [57].

In a recent systematic review, 53% of apps offered documentation function (recording and displaying data), 17.8% analysis function (the possibility to analyze the recorded data and to graphically display the results), 11.4% reminder function (reminds the user of its periodic, predefined medication), 34.5% of apps offered an information function (inform about the illness). Data forwarding/communication function (opportunity to send the recorded data) was present in 31.1% of apps. Surprisingly, only 8.8% of the diabetes apps provided an advisory function (use of the recorded data to create individualized advice) or any other kind of therapeutic support. Besides, the previously described functions, 14.5% of the apps included suggestions for recipes suitable for the needs of diabetics. The majority of apps, i.e., 54.1% were limited to only one function, while there were only 28.2% with two and 17.7% with three and more functions [58]. In another classification, features of apps were grouped into three classes on the basis of prevalence. In class A, there were insulin and medication management, communication and patient monitoring by primary care providers, diet management, and physical activity. Class B included weight management, blood pressure management, and connection to personal health record (PHR). In class C, there were education, social media, and alerts. Class A comprised four major features and class B had significantly higher prevalence than class C [59].

mHealth research platform *Few Touch Application* (FTA) was developed to support manage‐ ment of diabetes. Applications and studies based on FTA allow automatic monitoring of blood glucose information, receiving short message service information about type 1 diabetes, mobile diary for type 2 diabetes, sharing diaries with doctor or nurse, mobile diary for type 1 diabetes, a food picture data, transfer of physical activity data on mobile phone, nutrition advices, context sensitivity, and modeling of blood glucose. Performance of each of the 10 FTA-based apps was analyzed and the conclusion was that all FTA apps are beneficial [60].

In the next sections, we will present 11 problems of diabetes disease that can be addressed by mHealth.

#### **7.1. Glycemic control**

**6. Why mobile health?**

36 Mobile Health Technologies - Theories and Applications

management.

of them.

**7. Features of diabetes apps**

Mobile health is defined as a use of mobile communication devices, which include mobile phones, patient monitoring devices, tablets, personal digital assistants, and other wireless devices for health services, and information [52]. Currently, growing use of those devices can be seen in practically every part of the world. Number of mobile phone owners in the United States has reached 92% of a population and even number of smartphone owners has grown to 68%, while 45% of people own a tablet computer. Desktop or laptop computers were bought by 73% of Americans [53]. Surprisingly, similar rise in mobile technology use is also recorded among people in developing countries, where average share of people with mobile phones is around 83% [54]. With a vast majority of world population having an access to some type of mobile technology, this can certainly become a widely used to deliver deferent health care solutions to people. Role of mHealth is very broad and includes education and rising aware‐ ness, remote data collection, remote monitoring, communication with health care workers, support with diagnostic and treatment, and tracking diseases and epidemic outbreaks [55]. Most of these tasks are already performed by various mHealth applications for diabetes self-

There are different classifications of diabetes app features and in this section we present some

Review of accessible diabetes applications has shown that they mostly focus on blood glucose monitoring, medications, physical exercise, and diet management, while they also include other features such as education, communication, weight or BMI and blood pressure tracking, integration with public health records, decision support systems, and social networking. Blood glucose monitoring is available in all reviewed applications, while other features are more rarely present. Educational tools are brought to use in just 18% of applications and only around 30% of applications offer means to monitor weight, blood pressure, and physical exercise [56]. Still, among all medical conditions, diabetes with weight control represents the most addressed

In a recent systematic review, 53% of apps offered documentation function (recording and displaying data), 17.8% analysis function (the possibility to analyze the recorded data and to graphically display the results), 11.4% reminder function (reminds the user of its periodic, predefined medication), 34.5% of apps offered an information function (inform about the illness). Data forwarding/communication function (opportunity to send the recorded data) was present in 31.1% of apps. Surprisingly, only 8.8% of the diabetes apps provided an advisory function (use of the recorded data to create individualized advice) or any other kind of therapeutic support. Besides, the previously described functions, 14.5% of the apps included suggestions for recipes suitable for the needs of diabetics. The majority of apps, i.e., 54.1% were limited to only one function, while there were only 28.2% with two and 17.7% with three and

medical issue by mHealth applications in mHealth research [57].

Monitoring of blood glucose level is a base function of all available mHealth diabetes appli‐ cations, because even without technology interventions self-monitoring of blood glucose (SMBG) is still an integral component of daily diabetes management, especially for insulintreated patients [56, 61]. Unstructured SMBG is not recommended and does not produce the same results as structured SMBG, which links to behavioral changes, optimization of therapy, and improved clinical outcome. An example of such structured SMBG is a 7-point glucose profile, where blood glucose is measured every day of the week preprandially, postprandially, and at bedtime [62]. A pilot of such structured SMBG demonstrated a reduction in HbA1c levels up to 1.2% in 12 months [63]. To help patients in keeping up with structured SMBG, mHealth offers personal goal setting and various types of reminders [64]. Most commercially available devices for glucose monitoring enable patients to store and follow their blood glucose pattern. For this, patients need to transfer measurements to a computer through an USB cable or to a mobile phone with a direct connection through Bluetooth or Wi-Fi. Even more accurate glucose profile can be obtained with a use of continuous glucose monitor [61]. Regardless of a type of a mHealth intervention used, there is evidence in its positive impacts on reduction of HbA1c values by a mean of 0.5% over 6 months [65]. However, a review of 24 papers has shown that the effectiveness of mHealth interventions (blood glucose reading and transmis‐ sion to server) measured in HbA1c value was inconsistent for both types of diabetes [66].

#### **7.2. Nutrition control**

For education, most of the technologies used for nutrition therapy rely on videoconferencing, while for food tracking and food selection various mobile apps [47]. Distinct features are being formed among behavior mHealth modalities. Food intake can be recorded to determine the quantity of calories consumed each day and the targeted quantity of calories automatically adjusted, based on patient's daily physical activities [67]. More widely adopted are different nutrition databases containing a rage of food items, including different brands and restaurant food, and real-time calculations of consumed calories. Similarly, there exists also a possibility to scan the barcode of brand names to see the nutrient content [47]. Growing number of possibilities provide a new generation of mHealth devices also known as wearables. For example, in diet self-management wrist monitors and electronic utensils can be used to track the amount and speed of bites, but such devices are practically not used yet [68]. Furthermore, mHealth may enable calories calculations with recognition of food from photography in freeliving conditions. Even more promising are mobile applications that suggest appropriate meal based on preprandial blood glucose reading, which can facilitate patients' educated decision making [64].

#### **7.3. Physical activity**

Sixty nine percent of diabetes patients describe their exercise practices as nonexistent or less than recommended level [69]. It is recommended that adult patient with diabetes perform at least 150 min of moderate-intensity aerobic physical activity per week, spread over at least 3 days and with no more than two consecutive days without exercise [70]. For patients to monitor the extent of their daily physical activity, mHealth offers solutions in a form of body-worn activity monitors. Most easily accessible are pedometers, but besides number of steps taken they do not measure other forms of physical activity [64]. Meanwhile, accelerometers with a combination of gyroscope can record wider range of movements and accuracy of recordings is not dependent on person's body position [71]. When tested, people with access to fully automated system performed on average for 2 h and 18 min per week more of physical activity than people without it [72]. Wearable sensors still need to be complemented with education, planning, and feedback tools to successfully promote physical activity. Effectiveness of mHealth intervention was shown in improved daily number of steps, which was done by setting an achievable goal, providing real-time feedback about the amount of burned calories, and showing recorded progress. This raised the number of steps by 22% in 8 weeks [73].

It was observed that insufficient number of currently evaluable mHealth applications incor‐ porate evidence-based behavior change techniques. This is especially true for techniques, such as relapse prevention, teaching to use props, time management, and agreement to form behavior contract [74].

#### **7.4. Weight loss**

Obesity should be diagnosed according to body mass index (BMI). BMI classes are normal weight (18.5–24.9 kg/m2 ), overweight (25–29.9 kg/m2 ), obesity class I (30–34.9 kg/m2 ), obesity class II (35–39.9 kg/m2 ), and obesity class III (≥40 kg/m2 ). For Southeast Asians and Asian Indians, lower BMI cut-points may be appropriate. Lifestyle modifications including behav‐ ioral changes, reduced calorie diets, and appropriately prescribed physical activity should be implemented as the cornerstone of obesity management [75]. Weight loss can be achieved with 500–750 kcal/day reduction that means intake of 1200–1800 kcal/day depending on sex and baseline body weight [76]. Raised BMI, i.e., above 25 kg/m2 , is seen in more than 75% of diabetes patients [77]. Patients with high BMI and diabetes are significantly more likely to have poor glycemic control [78]. Overweight individuals with diabetes are encouraged to lose at least 5– 10% of their weight, because this importantly reduces most cardiovascular risk factors, but it is worth mentioning that lager weight loss (10–15%) has even greater benefits [79]. Reviewed mHealth applications offer means to help achieve this recommendation by monitoring and facilitating physical activity (41% of the applications) and by improving users' diet (68% of the applications) [56]. An evaluation of 137 diabetes apps showed, that only 39% of them offered weight tracking [59]. Self-monitoring of weight and of body composition by using weight scales can now be accomplished wirelessly with mHealth apps or computer applications. This minimizes the burden on the user, while it also minimizes the error in data transcription. Tracked weight and fat mass can be graphically analyzed by the patient or health care practitioner [47].

SMS interventions were investigated to promote change in diet and physical activity. Small and short randomized controlled trials proved significant weight loss, while larger and longer studies showed no statistical significance [80].

Researchers investigated dietary self-monitoring-based electronic interventions using person‐ al digital assistants (PDAs), electronic portable devices that share some of the features of mobile phones. PDA in the study was equipped with dietary and exercise software with and without feedback message. Patients were enrolled in three groups: PDA alone, PDA with feedback (feedback algorithm that provided daily messages tailored to their entries and provided positive reinforcement and guidance for goal attainment), and paper diary/record. All participants had statistically significant weight loss, but PDA group combined with feedback had the highest proportion of participants achieving greater than 5% weight loss in six months [81].

Studies incorporating podcasts compared to podcasts that included prompting by mobile app and interaction with study counselors and other participants on Twitter, did not show enhanced weight loss in the latter group [82].

Interventions delivered by smartphone app, website, or paper diary were also compared. App incorporated goal settings, self-monitoring of diet and activity, and feedback via weekly text message. The website group used commercially available slimming website. Trial retention, adherence, and achieved weight loss were significantly higher in the smartphone app group [80].

#### **7.5. Blood pressure control**

quantity of calories consumed each day and the targeted quantity of calories automatically adjusted, based on patient's daily physical activities [67]. More widely adopted are different nutrition databases containing a rage of food items, including different brands and restaurant food, and real-time calculations of consumed calories. Similarly, there exists also a possibility to scan the barcode of brand names to see the nutrient content [47]. Growing number of possibilities provide a new generation of mHealth devices also known as wearables. For example, in diet self-management wrist monitors and electronic utensils can be used to track the amount and speed of bites, but such devices are practically not used yet [68]. Furthermore, mHealth may enable calories calculations with recognition of food from photography in freeliving conditions. Even more promising are mobile applications that suggest appropriate meal based on preprandial blood glucose reading, which can facilitate patients' educated decision

Sixty nine percent of diabetes patients describe their exercise practices as nonexistent or less than recommended level [69]. It is recommended that adult patient with diabetes perform at least 150 min of moderate-intensity aerobic physical activity per week, spread over at least 3 days and with no more than two consecutive days without exercise [70]. For patients to monitor the extent of their daily physical activity, mHealth offers solutions in a form of body-worn activity monitors. Most easily accessible are pedometers, but besides number of steps taken they do not measure other forms of physical activity [64]. Meanwhile, accelerometers with a combination of gyroscope can record wider range of movements and accuracy of recordings is not dependent on person's body position [71]. When tested, people with access to fully automated system performed on average for 2 h and 18 min per week more of physical activity than people without it [72]. Wearable sensors still need to be complemented with education, planning, and feedback tools to successfully promote physical activity. Effectiveness of mHealth intervention was shown in improved daily number of steps, which was done by setting an achievable goal, providing real-time feedback about the amount of burned calories, and showing recorded progress. This raised the number of steps by 22% in 8 weeks [73].

It was observed that insufficient number of currently evaluable mHealth applications incor‐ porate evidence-based behavior change techniques. This is especially true for techniques, such as relapse prevention, teaching to use props, time management, and agreement to form

Obesity should be diagnosed according to body mass index (BMI). BMI classes are normal

Indians, lower BMI cut-points may be appropriate. Lifestyle modifications including behav‐ ioral changes, reduced calorie diets, and appropriately prescribed physical activity should be implemented as the cornerstone of obesity management [75]. Weight loss can be achieved with 500–750 kcal/day reduction that means intake of 1200–1800 kcal/day depending on sex and

), obesity class I (30–34.9 kg/m2

). For Southeast Asians and Asian

), obesity

), overweight (25–29.9 kg/m2

), and obesity class III (≥40 kg/m2

making [64].

**7.3. Physical activity**

38 Mobile Health Technologies - Theories and Applications

behavior contract [74].

weight (18.5–24.9 kg/m2

class II (35–39.9 kg/m2

**7.4. Weight loss**

Elevated blood pressure is a common condition coexisting with diabetes and it is a clear independent risk factor for the cardiovascular complications. To reduce the risk, blood pressure should be routinely monitored and maintained at a targeted level. Recommended systolic pressure for diabetes patient is <140 mmHg and diastolic pressure <90 mmHg. Homemonitoring is greatly encouraged, because it is a way to exclude white coat hypertension and because research suggested better correlation between at home measurements and cardiovas‐ cular risk than office measurements [83]. Among reviewed diabetes applications, 23% of them currently offer means of self-monitoring of blood pressure [56]. Monitoring blood pressure with the help of mHealth program may detect hypertension patterns that would otherwise gone unreported [22]. Fully automatized blood pressure readings, which are immediately stored in mHealth device and send to a health care provider, enabled to 51% of diabetes patient to reach the targeted blood pressure level. This is a significant improvement compared to 37.5% of general diabetes population that succeed in lowering their blood pressure. Such improve‐ ment was achieved also due to by inclusion of daily reminders, alerts in a case of concernreassign measurement and linkage to patient support system [84]. Home-monitoring alone does not produce the same results, proving telemedicine equal, but not more effective than standard approach, so this needs to be taken into consideration for future mHealth design [85].

#### **7.6. Medication adherence**

Previously, already discussed lack of adherence to pharmacological intervention in diabetes patients is an issue addressed by approximately 76% of reviewed mHealth applications [56]. One of the commonly reported barriers is patient forgetfulness, when it comes to medication intake regime. Mobile health technologies can offer different solution to this problem. Daily automatic electronic or text-message reminders may improve medication intake [86]. Those reminders can be upgraded by the use of real-time medication monitoring, which is possible with a use of electronic medication dispenser that records date and time of each opening. Consequentially, alert is only send in a case of forgotten medication dose. A trail confirmed that a baseline adherence of around 62% improved to 79.5% adherence after 1 year of use. Long-term effectiveness of this mHealth method peaked at 80.4% medication adherence, whereas control group adherence remained in a range of 68.4%. Majority of patients that used real-time medication monitoring also agreed that this method supported higher awareness of their medication use and reported positive experience with receiving SMS reminders [87, 88]. Considerable amount of patient, who tested real-time monitoring devices, were glad that physician knew if they took their medication and were reassured by technology supported communication [89].

#### **7.7. Education**

Diabetes education is a key element in patient care. To reassure adequate results, effective education strategies can be found in National Standards for Diabetes Self-Management Education and are worth applying to mHealth methods [90]. Even limited amount of education can result in improved weight control and potentially reduced cardiovascular risk [91]. Initial comparisons between in person diabetes education and education administrated through telemedicine already demonstrated a feasibility and equal effectiveness of technology sup‐ ported methods [92]. Most diabetes self-management applications do not integrate educational information. When available, such information is often generic and is not personalized to the individual patient. This is more prominent in commercial applications [56]. Education and personalized feedback are still underdeveloped features, included in less than one third of reviewed mHealth applications. Only 20% of reviewed applications had an education module, and only 26% of these met the criteria for personalized education or feedback. Task of personalizing rapidly growing number of information is challenging, but it may be largely beneficial for diabetes patient [59]. Most widely used mHealth method for diabetes education is SMS. Meta-analysis of current findings has shown that mobile SMS education can improve glycemic control. The glycemic control is even better if diabetes education is done by a combination of SMS and internet methods, i.e., 86% effect in comparison with 44% that is achieved by SMS alone [93]. Positive results of e-mail and SMS education can also be seen in improved quality of life [94].

Numerous applications are available helping healthy people or people with risk factors to assess their risk for developing diabetes type 2 in the future. Only a few of these apps disclose the name of the risk calculator used for assessing the risk of diabetes, therefore the quality of their calculations is questionable [95].

#### **7.8. Diabetic retinopathy screening**

cular risk than office measurements [83]. Among reviewed diabetes applications, 23% of them currently offer means of self-monitoring of blood pressure [56]. Monitoring blood pressure with the help of mHealth program may detect hypertension patterns that would otherwise gone unreported [22]. Fully automatized blood pressure readings, which are immediately stored in mHealth device and send to a health care provider, enabled to 51% of diabetes patient to reach the targeted blood pressure level. This is a significant improvement compared to 37.5% of general diabetes population that succeed in lowering their blood pressure. Such improve‐ ment was achieved also due to by inclusion of daily reminders, alerts in a case of concernreassign measurement and linkage to patient support system [84]. Home-monitoring alone does not produce the same results, proving telemedicine equal, but not more effective than standard approach, so this needs to be taken into consideration for future mHealth design [85].

Previously, already discussed lack of adherence to pharmacological intervention in diabetes patients is an issue addressed by approximately 76% of reviewed mHealth applications [56]. One of the commonly reported barriers is patient forgetfulness, when it comes to medication intake regime. Mobile health technologies can offer different solution to this problem. Daily automatic electronic or text-message reminders may improve medication intake [86]. Those reminders can be upgraded by the use of real-time medication monitoring, which is possible with a use of electronic medication dispenser that records date and time of each opening. Consequentially, alert is only send in a case of forgotten medication dose. A trail confirmed that a baseline adherence of around 62% improved to 79.5% adherence after 1 year of use. Long-term effectiveness of this mHealth method peaked at 80.4% medication adherence, whereas control group adherence remained in a range of 68.4%. Majority of patients that used real-time medication monitoring also agreed that this method supported higher awareness of their medication use and reported positive experience with receiving SMS reminders [87, 88]. Considerable amount of patient, who tested real-time monitoring devices, were glad that physician knew if they took their medication and were reassured by technology supported

Diabetes education is a key element in patient care. To reassure adequate results, effective education strategies can be found in National Standards for Diabetes Self-Management Education and are worth applying to mHealth methods [90]. Even limited amount of education can result in improved weight control and potentially reduced cardiovascular risk [91]. Initial comparisons between in person diabetes education and education administrated through telemedicine already demonstrated a feasibility and equal effectiveness of technology sup‐ ported methods [92]. Most diabetes self-management applications do not integrate educational information. When available, such information is often generic and is not personalized to the individual patient. This is more prominent in commercial applications [56]. Education and personalized feedback are still underdeveloped features, included in less than one third of reviewed mHealth applications. Only 20% of reviewed applications had an education module,

**7.6. Medication adherence**

40 Mobile Health Technologies - Theories and Applications

communication [89].

**7.7. Education**

Diabetic retinopathy represents most frequent cause of newly occurring adult blindness. Incidence of diabetic retinopathy is highly depended on duration of diabetes itself. Among population with type 2 diabetes cumulative incidence after 4 years is estimated to be 26%, 38.1– 41.0% after 6 years and 66% after 10 years [96]. Comprehensive eye examination should be performed after diabetes diagnosis and repeated every 2 years, if there are no visible changes, or annually, if initial retinopathy changes are already present [97]. To keep up with this requirement even with patient in remote and isolated areas, low cost smartphone-based intelligent system integrated with microscopic lens was developed and tested. System detects retinal abnormalities by a method of comparison with medical image database. Early testing has promisingly shown more than 87% accuracy rate in retinal disease detection [98].

#### **7.9. Diabetic foot screening**

Diabetic foot is a main cause for nontraumatic lower limb amputation and precedes 85% of the cases. Approximately 15% of diabetes patients will develop diabetic foot ulcers in their lifetime [99]. It is recommended for all diabetes patients to perform annual extensive foot examination to identify risk factors predictive of foot ulcers. Patient should be screened for signs of peripheral neuropathy and peripheral vascular disease, simultaneously paying attention to identify other risk factors such as foot deformities, past foot ulcers, visual impairment, and cigarette smoking [97]. Currently, researched mHealth method to facilitate foot care is using high quality photography to diagnose foot ulcers and preulcerative lesions. Trained profes‐ sional can diagnose visible diabetic foot changes in valid and reliable manner, which implies methods usability as a monitoring tool in home environment [100]. Originally developed was a method for wound area measurement. The wound margins are recognized with the help of smartphone. First, the wound contour is copied on the foil. The area is then measured with smartphone app and compared with previous measurements [101].

#### **7.10. Psychosocial care**

Patients with diabetes should regularly be assessed for their psychosocial well-being. Care should include assessment of their attitudes about the illness and prognosis, mood changes, satisfaction with quality of life, financial, social and emotional conditions, and possible psychiatric disorders (depression, distress, anxiety, eating disorders, dementia). Screening is recommended for depression and cognitive impairment for older than 65 [76].

Telemedicine study researching depression and glycemic control in elderly showed a weak relationship between depression and HbA1C, but depression did not prospectively predict change in glycemic control [102]. In another study, web-based depression treatment for diabetics using cognitive behavior therapy was effective in reducing depressive symptoms, diabetes-specific emotional distress, while it had no benefit on glycemic control [103].

Telephone-based cognitive assessment (TBCA) was previously performed by conventional telephone. Because of better understanding of cognitive impairment, there is a possibility of more accurate TBCA. It needs more complex features of telephone which are easily achieved with the use of smartphones [104].

#### **7.11. Personal health record (PHR)/electronic medical record (EMR) and social media integration**

PHR is an internet-based set of tools that allows people to access and coordinate their lifelong health information and make appropriate parts of it available to those who need it [105]. Electronic medical record (EMR) is an electronic record with documents of patient's treatment in a clinic. Electronic health record (EHR) is a summary of individual's lifetime health status and care. Terms EMR and EHR are often used interchangeably [106].

Overall, 21% of commercial applications support synchronization of data with PHRs. Half of reviewed studies have integrated PHR with EMR and provide both patient and physician Web portal, whereas other included either a patient view or a clinician view of the EMR [56].

Social media integration is also emerging function of diabetes apps. It can help patients find similar users and communities in a dynamic fashion. But the majority of apps only provides links to their groups in well-known social networking sites such as Facebook and Twitter or maybe provides an account to a forum [59].
