Interview results on social support and diabetes app use (Study 1)
As a foundation for comprehensively examining the influence of (H1) professional and (H2) non-professional support on diabetes app use, and to specify the hypotheses, themes were extracted from the interview data, relating to the aspects "previous app use for diabetes management", "professional support", and "non-professional support". Table 2 summarizes the results in form of a diabetes app (non-) user typology, which displays the differences in social support in differing app user and non-user groups.
Diabetes app (non-) use in the sample – description of the dependent variable: App (non-) use categories ranged from "no previous use", "(no) interest in apps", and "(no) knowledge about existing diabetes management apps", to "infrequent and short-term app use" to "long-term app use". Some participants had never used diabetes apps for self-management before and expressed no interest in them (e.g., Jie, Kang), while others showed interest but lacked knowledge about an appropriate app use (e.g., Li Ting, Zhen Wei). Most participants were familiar with available diabetes apps (apart from a few without any knowledge about app availability, e.g., Rei Hong and Zhen Wei). Reported diabetes-specific app use was almost exclusively limited to logbooks for blood glucose monitoring (e.g., DAFNE online App, MySugr, Glooko, DiabetesM) and to food databases displaying nutritional information (e.g., food database app developed by the Singaporean Health Promotion Board). Moreover, app users split into short-term users ("adopters") who were mostly unsatisfied with the current state of diabetes apps and had abandoned their use after a while (e.g., Adit, Cheng), users who constantly switched apps, using several in parallel (e.g., Bharat, Navin), and users who used one main app over a longer period of time (e.g., Gu Fang, Henna) (Table 2).
Professional support – Physician-patient relationship: The themes derived from the interviews regarding healthcare professional support in (technology-supported) diabetes management included "medical specialty of physician related to perceived care quality", "taking time for communication", "actual decision-making", "decision-making preferences", and "inclusion of apps in physician-patient relationship".
Physicians were mentioned as the main supervisors in diabetes care, with other healthcare professionals (e.g., dieticians, nurse educators, podiatrists, pharmacists) only partly included in care with a large variation in respondents (Xin Qi, Bharat, Henna, Deng Li). Thus, we further focused on the physician-patient relationship.
Patients supervised by general practitioners reported very short consultations with short physician-patient communication ("the doctor is 5-10 minutes only, Xin Qi, age 56, T2DM; "if you ask questions, they will answer. But they won’t engage you for too long", Jie, age 64, T1DM), perceived lacking diabetes knowledge, and a perceived lack in support ("they [physicians] are not helpful", Li Ting, age 49, T2DM). They mostly expressed being unsatisfied with the quality of supervision by their general practitioners. The group of dependent patients, following doctors’ instructions and mainly not taking diabetes-related decisions or being less active in decision-making (e.g., Rei Hong, age 61, T2DM), mostly consulted general practitioners. Additionally, non-users of apps mostly consulted general practitioners in the sample (Table 2).
In contrast to general practitioners, diabetes specialists were reported to take more time for support: "you got two types [of physicians]... one we call it family physician… one uh he charge you more, double [specialist]…the doctor will spend more time" (Bharat, age 66, T2DM). Moreover, specialists sometimes developed close relationships with their patients ("he’s... more like a family friend ... than a doctor", Henna, age 60, T2DM) and mostly followed shared decision-making approaches. Yet, some patients preferred to take diabetes care decisions independently, and considered their physicians as advisors only (e.g., Sona, Bharat) (Table 2).
Overall, very few patients reported that physicians or other healthcare professionals (nurses) talked about diabetes apps in the consultations, or shared app information with them (e.g., Xin Qi, Bharat, Henna). Some patients participated in diabetes programs (e.g., DAFNE, dose adjustment for normal eating) that included specialist supervision and an app for self-management (e.g., Sona, Shi Hui). Apart from these programs, physicians hardly used apps to communicate with their patients (sometimes Email, e.g., Gu Fang).
Non-professional support – Family & friends' support: Self-management support by the family and by friends fell into the categories "managing alone", "negative support", "involvement only after diagnosis", "strong involvement". Both app users and non-users reported receiving support by their family and friends in their diabetes management (Table 2), either after diagnosis at the beginning of self-care (e.g., Cheng, Pang) or throughout the whole process of self-care (e.g., Navin, Xiu Wen). Negative support was reported when the family or friends pushed the patients towards unhealthy lifestyles, e.g., "They always say 'never mind la! Eat la, just eat la. Only once. You don’t eat this every day'" (Kang, age 67, pre-diabetes), or when an involvement of the family resulted in nagging behaviors ("whenever they are with me... when the doctor tells me something… and then after … when we go home they immediately start nagging me", Adit, age 22, T1DM). Positive attitudes towards diabetes care by the family/friends were mentioned as important for self-management ("the key is that they're... not uhh... really that negative on this... they're also very positive", Xiu Wen, age 57, T1DM). Some non-users of diabetes apps said they managed their condition alone, without the involvement of family or friends ("It’s myself, nobody else... No friends, no, nobody else, it is me", Jie, age 64, T1DM).
Overall, the interviews showed that aspects of professional support (style of decision-making, duration and quality of communication) related to perceptions of satisfaction and success in self-management. Moreover, this appeared to be influenced by the medical specialty of the physician supervising the patient (compare 17). Regarding non-professional support, the support by family and friends diversely related to (technology-supported) self-care, with both positive and negative influences of the family/friends on self-management reported. Based on the exploratory interview results we specified the theoretically derived hypotheses H1 and H2 as following:
(H1a) Supportive behaviors by the supervising physician (shared decision-making styles and supportive communication) positively predict diabetes app use for self-management.
(H1b) The medical specialty of the supervising physician (specialist versus general practitioner) is a predictor of diabetes app use for self-management, with specialist care promoting app use to a larger extent than care by general practitioners.
(H2a) (Positive) support by the personal social patient network (family and friends) positively predicts diabetes app use for self-management.
Online survey results on social support and diabetes app use (Study 2)
In Study 2 we tested these hypotheses (H1a), (H1b) and (H2a) using binary logistic regression. Checking for autocorrelation of all independent variables, physician decision-making and physician-patient communication were highly correlated with r = .772, p < .01. Therefore, decision-making and communication were recoded into a single variable “physician-patient relationship” (17).
Based on theoretical considerations and the interview results, we started with a binary logistic regression model that included a maximum of independent factors: physician-patient relationship, medical specialty of the physician, family/friend support, and other relevant factors from technology adoption theory (50) and from self-management theory (5) that had been shown to be relevant predictors of technology use for disease self-management. We then compared different models by reducing independent factors to find the model with the best fit.
A model with good fit included the factors derived from the interviews physician-patient relationship, family/friend support, medical specialty of the physician, as well as the additional UTAUT factors’ perceived app potential (performance and effort expectancy), previous health information seeking online (technological experience), and age; and the self-management factors type of diabetes, length of diabetes, perceived health status, payment problems and insurance coverage (practical socioeconomic), blood glucose testing adherence (self-management behaviors), interest in innovation (attitudes), perceived diabetes knowledge, program or support group participation, and psychological empowerment. The test of this model against the constant-only model was statistically significant, indicating that the predictors as a set reliably distinguished between diabetes app use and non-use (χ2(16) = 26.752, p < .05). Nagelkerke’s R2 = .656 indicated a high relationship between prediction and grouping (goodness-of-fit). Prediction success overall was 90% (80% for app non-use and 96% for app use). The Wald criterion demonstrated that only the family/friend support (Wald(1) = 5.315, p = .021) and the medical specialty of the consulted physician (dummy general practitioner or specialist, Wald(1) = 4.014, p = .045) made a significant contribution to the prediction of diabetes app use. The Exp(β) value indicated that when the family/friend support was increased, the relative probability (odds ratio) that diabetes apps were used decreased with Exp(β) = .044, β = -3.131. The Exp(β) value indicated that when the patients were supervised by specialist doctors the relative probability (odds ratio) that diabetes apps were used increased with Exp(β) = 9460.805, β = 9.155. In contrast to non-professional support (family/friends), the physician-patient relationship was not found to be a significant predictor in the model.
A model resulting after the removal of the factors interest in innovation, insurance coverage, and program or support group participation (due to lacking significance) showed a slightly lower prediction success with 75% (70% for app non-use and 79% for app use), but overall model significance with χ2(13) = 26.936, p < .05, and a moderate to high relationship between prediction and grouping (goodness of fit) with Nagelkerke’s R2 = .509. In this model, the Wald criterion demonstrated that the family/friend support (Wald(1) = 6.617, p = .010) and the perceived health status (Wald(1) = 7.839, p = .005) made a significant contribution to the prediction of diabetes app use. Again, the Exp(β) value showed that when support by family/friends was increased, the relative probability (odds ratio) that diabetes apps were used decreased with Exp(β) = .283, β = -1.261. The Exp(β) value also indicated that when the perceived health status was improved the relative probability (odds ratio) that diabetes apps were used increased with Exp(β) = 8.030, β = 2.083.
Further reducing the independent factors, the models showed similar results to the last model, resulting in family/friend support and the perceived health status being significant predictors of diabetes app use. A further reduction of factors decreased the model fit, yet the medical specialty of the physician nearly reached significance again.
Overall, after testing various models, only the family/friend support, the medical specialty of the supervising physician, and the perceived health status significantly predicted diabetes app use. Less family/friend support was likely leading to a higher chance of diabetes app use, while the use of diabetes specialists or a better perceived health status increased the chance of app use for self-management.
Despite the physician-patient relationship lacking significance for predicting diabetes app use in the models (apart from the medical specialty of the physician as a significant predictor), additional t-test calculations showed that the medical specialty of the physician related to the style of decision-making and the physician-patient communication. There were differences between specialists and general practitioners, with specialist care positively related to higher shared decision-making and better physician-patient communication as compared to general practitioners (Decision-making PDMstyle: specialists: M = 3.72, SD = 1.16, n = 37, general practitioners: M = 2.65, SD = 1.30, n = 22; t(57) = -3.275, p < .01, N = 59; Communication PCOM: specialists: M = 3.82, SD = 1.01, n = 37, general practitioners: M = 3.13, SD = 1.09, n = 22; t(57) = -2.472, p < .05, N = 59) (compare 17).
 all names have been changed to ensure anonymity