Lifestyle Monitoring and Coaching Using the Mobile DIAMETER Application in Secondary Care (DIAMETER-1)

Lifestyle Monitoring and Coaching Using the Mobile DIAMETER Application in Secondary Care (DIAMETER-1, Phase 2 - Feasibility Study).

Type 2 Diabetes Mellitus (T2DM) is a major chronic lifestyle-related disorder with a significant impact on quality and costs of care. As patients with T2DM often have insufficient knowledge about proper self-management and are insufficiently motivated for a lifestyle change, interventions with more motivational strategies and personalization are needed. The use of real-time monitoring of glucose values, nutrition and physical activity in combination with coaching aimed at lifestyle-related behavior change may improve patients' diabetes management. Therefore, ZGT, UT and RRD developed the Diameter app. The aim of phase 2 of this study is to investigate the Diameter as blended-care using a feasibility study. The primary objective of this feasibility study is to assess intervention usage and acceptability of the Diameter as a blended-care intervention in secondary care, of which some are also following a combined lifestyle intervention (GLI). Secondary objectives are to explore behavioral (e.g. physical activity), physiological (e.g. BMI), psychological (e.g. health-related quality of life) and clinical outcomes (e.g. glucose control, estimated HbA1c values). This study has a mixed-method design with 3 (regular participants) or 4 (participants who decide to follow the Combined Lifestyle Intervention (GLI) COOL next to the Diameter during the study period) data collection points. Patients will start with a two-week period of baseline measurements. Subsequently, patients will use the Diameter as a blended-care intervention for 10 weeks. The two-week measurement periods will be repeated twice (T1: week 13-14 and at T2: week 25-26). Between T1 and T2, patients will use a version of the Diameter without daily coaching messages. At T1 and T2, questionnaires will be administered, data on physical activity, food intake and glucose values will be logged, and blood and urine samples will be retrieved from regular care measurements. In addition, open-ended interviews will be performed with 10-15 patients at T1. For participants who also decided to follow the COOL program, some routinely collected measurements as part of the COOL program will be obtained from the patient record.

INTRODUCTION Type 2 Diabetes Mellitus (T2DM) is a major chronic lifestyle-related disorder with a significant impact on quality and costs of care. As patients with T2DM often have insufficient knowledge about proper self-management and are insufficiently motivated for lifestyle change, interventions with more motivational strategies and personalization are needed. The use of real-time monitoring of glucose values, nutrition and physical activity in combination with coaching aimed at lifestyle-related behavior change may improve patients' diabetes management. It is hypothesized that a technology-supported lifestyle intervention is effective and a step forward in T2DM management. In the Twente region of the Netherlands, a personalized treatment strategy to improve lifestyle in T2DM patients is being developed by providing coaching and feedback in daily life through a mobile application, the Diameter. The development of the Diameter is a collaboration between ZGT hospital, University of Twente and Roessingh Research and Development Institute. The Diameter monitors food intake, physical activity and glucose values; it gives individual patients and healthcare professionals insight into lifestyle, blood glucose levels, as well as into the effect of lifestyle behavior on glucose values in daily life. Furthermore, the Diameter offers evidence-based coaching aimed at improving lifestyle, i.e. physical activity and nutrition. The coaching content offered in the Diameter comprises the e-supporter which is developed by the UT, ZGT and TNO within the E-Manager project. It consists of goal setting, daily lifestyle coaching via short messages and a weekly exercise that supports people to achieve selected goals for a period of ten weeks. An innovative and unique coaching strategy is the feedback on blood glucose levels through continuous glucose monitoring (i.e., FreeStyle Libre). This continuous feedback is considered to be a very powerful tool for behavior modification and offers great potential. The development of the Diameter is an iterative process. Since 2017, the monitoring and coaching elements of the Diameter have been developed, extended and validated in several studies. In a blended-care setting, the Diameter combines face-to-face counselling of usual T2DM care with the Diameter and continuous glucose monitoring through the FreeStyle Libre. OBJECTIVE The Diameter as a blended-care intervention has only been evaluated in a small pilot study (phase 1) in ZGT hospital to test the proposed research protocol for phase 2 with T2DM patients treated at the outpatient clinic. Therefore, information about the intervention usage and acceptability are not yet available, as well as proper data for sample size calculation in future effectiveness testing. The next step is to implement and evaluate the Diameter in practice regarding intervention usage and acceptability (e.g. effort expectancy, pleasure), and to explore its effect on clinical outcomes (e.g. glycemic control), physiological (e.g. body composition), behavioural (e.g. physical activity), and psychological outcomes (e.g. health-related quality of life). Phase 2, which is registered in this trial system, concerns a feasibility study in ZGT hospital to evaluate intervention usage and acceptability of the Diameter as part of usual care. Usual care includes traditional diabetes care, possibly for a subgroup of participants, supplemented by the combined lifestyle intervention "Gecombineerde Leefstijl Interventie" (GLI) COOL which will be offered by ZGT to their T2DM patients. STUDY DESIGN The Diameter will be assessed using a mixed-methods prospective longitudinal design among 80 patients with T2DM. Approximately 20-40 of them will also follow the COOL program as part of usual care. During the study, 3 rounds of measurement will be performed. Additionally, for participants who started the COOL program next to the use of the Diameter, an extra measurement moment after 8 months (T3) is included. The data for this measurement will be collected as part of usual care and will be extracted from the participants' health record with the patients' consent. The measurements will be performed at the following time points: T0 (week 1-2): All participants will start with a two-week period of baseline measurements, including blinded glucose measurement (using the Freestyle Libre Pro IQ sensor), blinded nutrients intake measurement (using the Diameter app), a non-blinded physical activity measurement (using a Fitbit), other clinical and physiological outcomes and some questionnaires. Use of the Diameter (week 3-12): All study participants use the Diameter for 10 weeks to receive digital personalized coaching with a maximum of 2 coaching messages per day and one exercise per week about goal setting and achievement. Also, they use the Diameter to monitor their physical activity and nutrition. They receive 2 Freestyle Libre 2 sensors with which they can measure their glucose values continuously for in total 4 weeks, using the LibreLink app. T1 (week 13-14): All participants will perform a two-week period of measurements of all clinical, physiological, and behavioral outcomes, will fill in a number of questionnaires and 10-15 participants will be asked to participate in an interview about their acceptability of the Diameter. Use of the Diameter "light" (week 15-24): The participants will use a "light" version of the Diameter, consisting of all functionalities of the Diameter except the daily coaching messages as they will stop automatically after 10 weeks. They receive 2 Freestyle Libre 2 sensors with which they can measure their glucose values continuously for in total 4 weeks. T2 (week 25-26): All participants will again perform a two-week period of measurements of clinical, physiological and behavioral outcomes and will fill in some questionnaires. The follow-up measurement is performed after 6 months because it is assumed that there is a permanent change in behavior when the healthy behavior is maintained for a period of six months or longer. T3 (week 34): Only applicable for participants who are also following the COOL program. After 6 months of using the Diameter, participants who also follow COOL will continue their trajectory with the lifestyle coach in ZGT. In the context of the COOL program, after 8 months some measurements are taken that are part of regular care, consisting of medication use, HbA1c, BMI and waist circumference. These data will be obtained with consent of the participant from the patient record. No additional measurements will be conducted for this purpose. These data will be used as a final follow-up measure to assess the extent to which these participants can maintain any progress in clinical and physiological outcomes after 8 months which is the end of the basic part of the program that focuses on behavioral change. STUDY POPULATION In phase 2, patients will be recruited in ZGT hospital. The patient population will consist of 80 T2DM patients (18+, male and female) visiting the outpatient clinic for T2DM at ZGT hospital. Patients who are already participating or participated in the DIALECT cohort will be recruited. The remaining number of patients will be recruited from the outpatient clinic in ZGT. Some of the T2DM patients recruited from the outpatient clinic also follow the COOL program within ZGT in addition to receiving regular diabetes care. People who will participate in the COOL program and want to participate in the study will start using the Diameter at the beginning of the COOL program. Patients who use blood glucose-lowering medication independent of their gender and socio-economic status will be included. STUDY SETTING This research project will be conducted within the outpatient clinic for diabetic care in the ZGT hospital. Patients will participate in the pilot study for 5 weeks and in the feasibility study, phase 2, for half a year. Research appointments will be combined with regular outpatient clinic or COOL visits as much as possible. Currently, 17 healthcare professionals are working in this specific outpatient clinic. The team consists of 8 internists, 7 T2DM nurses, one nurse practitioner and one technical physician specialist together treating 2500 patients with T2DM a year. The current patient population consists of 450 T2DM patients, they are on average 63 years old with a mean duration of 11 years, and most of them are male (58%). Their average BMI is 32.9 ± 6.2; almost all patients (95%) have a BMI higher than 25. In addition to traditional diabetes care, ZGT starts to offer the Combined Lifestyle Intervention (Gecombineerde Leefstijl Interventie; GLI) COOL to their T2DM patients. During the COOL program in ZGT, a professional lifestyle coach who is also a diabetes nurse will counsel adults with T2DM treated in ZGT who are obese or at high risk of obesity to achieve a healthier lifestyle. COOL consists of a basic part that focuses on behavioral change (8 months) and a maintenance part (16 months). In the first 8 months of the COOL program, participants follow 8 group sessions of 90 minutes and 4 individual sessions (2 sessions of 60 minutes, 2 sessions of 45 minutes).

The primary study aim is to assess intervention usage and acceptability (e.g. effort expectancy, pleasure) regarding the Diameter integrated in regular secondary care and how acceptability is associated with intervention usage. Therefore, the study does not make use of a double blind randomized controlled trial as this design does not fit the study's primary purpose. Consequently, there will be no randomization and treatment allocation. To minimize the treatment effect at baseline, baseline measures will be blinded as much as possible, i.e., participants will not be able to get insight in their glucose values and nutrient intake. The physical activity data collected with the Fitbit cannot be blinded, so that the participants do have insight into their physical activity during the baseline measurements

Regular T2DM treatment in secondary care will be complemented with the Diameter: a mobile application on Android smartphones that enables continuous monitoring of nutrition (via food diary), physical activity (via activity tracker Fitbit and self-reported activities) and blood glucose values (via Freestyle libre 2 sensors). The Diameter also provides autonomous lifestyle coaching via daily coaching messages, short weekly e-mails and exercises aimed at goal achievement.

Regular T2DM treatment in secondary care will be complemented with the Diameter: a mobile application on Android smartphones that enables continuous monitoring of physical activity (via activity tracker Fitbit and self-reported activities) and nutrition (via food diary) and blood glucose values (via Freestyle Libre sensor). The Diameter also provides autonomous lifestyle coaching via daily coaching messages, short weekly e-mails and exercises aimed at goal achievement.

The way in which the intervention is actually used in terms of frequency (number of time used over time)

Acceptability will be assessed using an UTAUT2 questionnaire of 19 questions. The UTAUT2 determinants include performance expectancy, effort expectancy, pleasure/hedonic motivation, facilitating circumstances, design, technical issues and habit. The experiences with the Diameter will be assessed using multiple statements generated by the research team according to literature and previous used UTAUT2-model based questionnaires. Each statement will be scored on a 7-point Likert scale (1 = extremely disagree, 7 = extremely agree) where a higher score indicate positive experiences with the Diameter. The scores will be used to calculate individual sum scores per determinant of the UTAUT2 model. Open-ended interviews will be conducted with a subset of 10-15 participation to gain in-depth information on the acceptability, and perceived barriers and facilitating conditions for usage of the Diameter.

The amount of glucose-lowering medication usage will be assessed via electronic health record data and printed prescriptions from the pharmacy, checked with the patient.

The total number of T2D- related medication, includes glucose-lowering medication, medication for blood pressure and medication for cholesterol, will be assessed via electronic health record data and printed prescriptions from the pharmacy, checked with the patient.

Estimated HbA1c (eHbA1c) will be calculated from glucose measurements assessed with the Freestyle Libre using accepted algorithms. The eHbA1c decreases with a least 10% for patients who are above target value of 53 mmol/mol, independent from the treatment schedule, and no clinically relevant increase in blood glucose lowering medication dose.

Time In Range (TIR) is a new parameter to evaluate blood glucose control. The TIR represents the total time per day that the glucose value is between 3.9 and 10.0 mmol/L. The higher the TIR, the better the diabetes is managed during the day. Algorithms will be used to calculate the TIR based on the glucose values assessed with the Freestyle Libre.

Time Below Range (TBR) represents the total time per day that the glucose value is below 3.9 mmol/L. The TBR will be calculated with algorithms based on the glucose values assessed with the Freestyle Libre.

Time Above Range (TAR) represents the total time per day that the glucose value is above 10 mmol/L. The TAR will be calculated with algorithms based on the glucose values assessed with the Freestyle Libre.

Body composition (impedance) measuring of the fat, muscle and water percentage which will be determined by 2 non-invasive devices: Bodyscan® Quadstad 4000 and TANITA® BC418MA. Body Mass Index (BMI, (kg/m2)) will be calculated using the measurement of weight and height collected in the outpatient clinic. Waist and hip circumference will be determined with a tape measure to assess the distance around the smallest part of the waist and the distance around the largest part of the hips.

Body Mass Index (BMI, (kg/m2)) will be calculated using the measurement of weight and height collected in the outpatient clinic.

Waist and hip circumference will be determined with a tape measure to assess the distance around the smallest part of the waist and the distance around the largest part of the hips.

HbA1c (mmol/mol) will be obtained from the blood samples that are taken as part of regular care from the patient record.

Fasting glucose (mg/dL) will be obtained from the blood samples that are taken as part of regular care from the patient record.

Total cholesterol (mmol/L) will be obtained from the blood samples that are taken as part of regular care from the patient record.

HDL-cholesterol (mmol/L) will be obtained from the blood samples that are taken as part of regular care from the patient record.

LDL-cholesterol (mmol/L) will be obtained from the blood samples that are taken as part of regular care from the patient record.

Triglycerides (mmol/L) will be obtained from the blood samples that are taken as part of regular care from the patient record.

eGRF (ml/min/1,73 m2) will be obtained from the blood samples that are taken as part of regular care from the patient record.

Creatinine (mmol/L) will be obtained from morning void urine as part of regular care, which will be sent to the laboratory for analysis and storage. Morning void data will be derived from the patient record and used for analysis.

Albumin (mg/L) will be obtained from morning void urine as part of regular care, which will be sent to the laboratory for analysis and storage. Morning void data will be derived from the patient record and used for analysis.

Albumin/creat ratio (g/mol kr) will be obtained from morning void urine as part of regular care, which will be sent to the laboratory for analysis and storage. Morning void data will be derived from the patient record and used for analysis.

Blood pressure will be measured as part of regular care three times automatically every minute for 15 minutes in a supine position through the Dinamap®.

Peripheral neuropathy will be evaluated with the touch test as part of regular care. This test is performed by pricking the monofilament under the big, middle and small toes of both feet. If the patient does not feel 2 or more of these pricks, this is diagnosed as having peripheral neuropathy.

Handgrip strength will be determined with a hand grip strength test which is an extra measurement compared to regular care.

The presence of toe/foot/leg amputations will be queried during the anamnesis. Specifically, participants will be asked about the year of amputation(s), on which side and at what level (from DIP joint to hip joint) the amputation took place.

The presence of ulcers will be queried during the anamnesis and assessed during the physical examination. The participant will be asked about the time of onset, the side, location and the current treatment of the ulcer. During the physical examination, the ulcer will be classified based on the Texas classification (depth 0-3, infection/ischemia A-D).

Physical activity will be assessed using the Fitbit or using the manually entered log-data in the Diameter. Steps per minute assessed with de Fitbit will be used provide information on levels of physical activity. In addition, these data can be used toe calculate other activity parameters (e.g., moderate to vigorous physical activity bouts, sedentary behavior, step cadence and peak activity index). To this purpose, scripts in Matlab have previously been developed and tested extensively. Steps per minute and total minutes of moderate to vigorous per week will be used to analyse the intervention effects.

Nutritional behavior will be assessed using the food diary in the Diameter. The Diameter logs (number of days logged, and mean number of logs per day), the eaten products at that time and nutritional components (carbohydrates, fat, protein, sugar and kilocalories). To assess the intervention effects, the mean number of carbohydrates, fat, protein, sugar and kilocalories per day will be used. Participants are asked to fill in the food diary for 6 days fper measurement moment, as 6 days are required to be able to measure variety in carbohydrabe intake. It is expected that filling in the food diary will take 10 minutes per day. This will take participants approximately a total of 60 minutes spread over 6 days per measurement moment (T0, T1 and T2), which makes a total of 180 minutes.

Treatment adherence will be assessed using de Beliefs about Medicines Questionnaire (BMQ). The BMQ consists of 19 items and comprises 2 sections: the BMQ-Specific which assesses representations of medication prescribed for personal use. The score for each subscale is between 5 and 20. A low score on necessity means a low confidence in the necessity of the prescribed medication, a high score indicates the opposite. A low score on the concerns (concern) means a high confidence in the positive effects of the prescribed medication. A high score means the opposite. The BMQ-General which assesses beliefs about medicines in general. The score for each subscale is between 4 and 20. A low score on the negative consequences (harm) indicates a great confidence in the positive consequences of the use of medication in general. A high score means low confidence in the positive consequences of the use of medication.

Self-efficacy will be assessed with all 20 items of the Diabetes Management Self-Efficacy scale for patients with T2DM (DMSES) using a 5-point Likert scale (1 probably not - 5 definitely yes). The scale was developed based on the self-care activities these patients have to carry out in order to manage their diabetes. The scale focuses on nutrition specific and weight, nutrition general and medical treatment, physical exercise and blood sugar. The 2 questions about the use of oral medication will be made optional as these questions do not apply for patients who have a treatment schedule that includes insulin. A sum score wil be calculated summing all items (range 20-100).

Self-management using the Patient Activation Measure (PAM) a 100-point quantifiable scale determining patient engagement in healthcare. It is a 13-item instrument which assesses patient (or consumer) self-reported knowledge, skills and confidence for self-management of one's health or chronic condition. Based on the PAM score patients can be classified into 4 categories ranging from more passive patients who perceived to have minimal control (1) to having perceived to being absolutely capable of managing their disease (PAM 4).

Stages of change using the single item questionnaire Self-Assessment Scales (SAS) Stages of Change. It assess a person's adherence to lifestyle guidelines using 5-point Likert scale. The stage of change is assessed with the following single question: Are you physically active for more than 30 minutes a day / consciously engaged in a healthy diet at least 5 days a week? The 5 answer options are traceable to the 5 stages of change of the Transtheoretical Model, i.e., precontemplation, contemplation, preparation, action and maintenance.

Health-related quality of life using the EQ-5D-5L. The EQ-5D-5L is a generic instrument for describing and valuing health. It defines health in terms of 5 dimensions: mobility, self-care, usual activities, pain/comfort, and anxiety/depression. Each dimension has 5 response categories ranging from no problems (1) to extreme problems (5). Also the VAS is included in this questionnaire. The EQ-VAS records the patient's self-rated health on a vertical visual analogue scale, where the endpoints are labelled 'The best health you can imagine' and 'The worst health you can imagine'. The VAS can be used as a quantitative measure of health outcome that reflect the patient's own judgement.

Duration of type 2 diabetes diagnosis (years) will be collected during the visit to the outpatient clinic.

Educational background (no education, primary education, lower secondary education, upper secondary education, bachelor, masters) will be collected during the visit to the outpatient clinic.

Migration background will be collected during the visit to the outpatient clinic. The migration background is asked because T2DM is relatively more common among certain cultures than among others. After the study, the question about migration background will be used to assess whether the different cultures are sufficiently represented by the participants in this study. If participants do not want to answer this question, they have the option to choose "I don't want to say this".

Medical history will be collected from the electronic patient files. Medical history of interest are past potential diabetic related complications such as: coronary disease, myocardial infarction, cardiac catheterization, stroke or heart bypass.

Openness to the use of new technologies measured with single-question based on the innovation theory by Rogers. This question enables to characterize people regarding 5 domains: innovators, early adopters, early majority, late majority and laggards.

Health literacy will be assessed using the Set of Brief Screening Questions (SBSQ). The SBSQ is validated in Dutch and contains three questions with a 5-point Likert scale (0 never/ very sure - 4 always/not at all sure) as answer options. The higher the score, the better the literacy. Low literacy is defined as a mean score of 2 or less.

Inclusion Criteria: Male and female patients with type 2 diabetes and being treated in the outpatient clinic at ZGT; Being familiar with using an Android smartphone (version 8.0 or higher); Aged 18 years or older; Competent: person can understand and weigh up information provided by researcher and can understand what the concequences of participation are. Provide written informed consent. Exclusion Criteria: Dependence on renal replacement therapy; Severe general diseases or mental disorders making the participation in the study impossible; Drug abuse Insufficient mastery of the Dutch language.

van Rinsum C, Gerards S, Rutten G, Philippens N, Janssen E, Winkens B, van de Goor I, Kremers S. The Coaching on Lifestyle (CooL) Intervention for Overweight and Obesity: A Longitudinal Study into Participants' Lifestyle Changes. Int J Environ Res Public Health. 2018 Apr 4;15(4):680. doi: 10.3390/ijerph15040680.

Glanz, K., B.K. Rimer, and K. VIswanath, Health behavior and health education: Theory, research, and practice, 4th ed. Health behavior and health education: Theory, research, and practice, 4th ed., ed. K. Glanz, B.K. Rimer, and K. Viswanath. 2008, San Francisco, CA, US: Jossey-Bass

Gant, C.M., Opportunities for Improvement of Cardiovascular Risk Management in Patients with Type 2 Diabetes and Chronic Kidney Disease: Integrated Assessment of Lifestyle Habits and Pharmacological Intervention in Routine Clinical Care. 2018, Rijksuniversiteit Groningen.

Venkatesh, V., J.Y. Thong, and X. Xu, Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology. MIS quarterly, 2012: p. 157-178.

Gill, P.K.S., et al., Handgrip strength in patients with type 2 diabetes mellitus. Pakistan Journal of Physiology, 2016. 12(2): p. 19-21.

Cetinus E, Buyukbese MA, Uzel M, Ekerbicer H, Karaoguz A. Hand grip strength in patients with type 2 diabetes mellitus. Diabetes Res Clin Pract. 2005 Dec;70(3):278-86. doi: 10.1016/j.diabres.2005.03.028.

Palaniappan U, Cue RI, Payette H, Gray-Donald K. Implications of day-to-day variability on measurements of usual food and nutrient intakes. J Nutr. 2003 Jan;133(1):232-5. doi: 10.1093/jn/133.1.232.

Baranowski, T., 24-hour recall and diet record methods. Nutritional epidemiology, 2012. 40: p. 49-69.

Horne, R., J. Weinman, and M. Hankins, The beliefs about medicines questionnaire: the development and evaluation of a new method for assessing the cognitive representation of medication. Psychology and health, 1999. 14(1): p. 1-24.

Bijl JV, Poelgeest-Eeltink AV, Shortridge-Baggett L. The psychometric properties of the diabetes management self-efficacy scale for patients with type 2 diabetes mellitus. J Adv Nurs. 1999 Aug;30(2):352-9. doi: 10.1046/j.1365-2648.1999.01077.x.

Rademakers J, Nijman J, van der Hoek L, Heijmans M, Rijken M. Measuring patient activation in The Netherlands: translation and validation of the American short form Patient Activation Measure (PAM13). BMC Public Health. 2012 Jul 31;12:577. doi: 10.1186/1471-2458-12-577.

Marttila J, Nupponen R. Assessing stage of change for physical activity: how congruent are parallel methods? Health Educ Res. 2003 Aug;18(4):419-28. doi: 10.1093/her/cyf034.

Herdman M, Gudex C, Lloyd A, Janssen M, Kind P, Parkin D, Bonsel G, Badia X. Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Qual Life Res. 2011 Dec;20(10):1727-36. doi: 10.1007/s11136-011-9903-x. Epub 2011 Apr 9.

Fransen MP, Van Schaik TM, Twickler TB, Essink-Bot ML. Applicability of internationally available health literacy measures in the Netherlands. J Health Commun. 2011;16 Suppl 3:134-49. doi: 10.1080/10810730.2011.604383.