As previously described [27], the recruitment of patients and documentation of data for this retrospective and prospective, observational, digital, patient-centered cohort study were carried out without the assistance of a health care professional (HCP), as part of the routine use of the MyTherapy app (smartpatient GmbH) by patients with T2DM in Germany. The app helps patients manage their treatment-related tasks and access information on other health topics, including comorbidities, by using their mobile device. The app provides the user with a daily “To do” list of tasks, including those regarding medication intake, and asks the user to confirm whether each task has been completed [27]. Patients who documented the following medications or medical devices within the app were considered eligible for participation during initial selection: metformin, acarbose, dipeptidyl-peptidase-4 inhibitor, glucagon-like peptide-1 receptor agonist, renin-angiotensin system inhibitor, SGLT2i, basal insulin, nonsteroidal mineralocorticoid receptor antagonist, sulfonylurea, and blood glucose test strips.

To ensure that patients were properly informed, all eligible patients who actively used the app and allowed push notifications during the app’s regular operation received content via push notifications during an awareness phase, which lasted from August 2 to 15, 2023. This content comprised 3 videos about the scope and goals of the SMART-Finder study, the risk of CKD in patients with T2DM, and the impairment of kidney function in patients with CKD; a study description page; a document for collecting laboratory values; and an electronic consent form for explaining, in detail, which data were used for this study and obtaining consent for the use of pseudonymized data. After completion of the consent form, an electronic case report form (eCRF), the EQ-5D-5L, and the Diabetes Treatment Satisfaction Questionnaire (DTSQ) were provided in the “Discover” section of the app (patients were notified about this via push notifications as well) on August 18, 21, and 24, 2023, respectively. The eCRF included questions on patients’ residential region, age, gender (self-reported as 1 of 3 options: male, female, or diverse), weight, height, smoking status, last systolic and diastolic blood pressures, serum fasting glucose level, and most recent available laboratory values within the last 12 months (UACR, eGFR, and glycated hemoglobin [HbA_1c]). If patients did not know these laboratory values, they were encouraged to request them from their treating physician.

The observation period for the baseline analysis was from August 18, 2023, until the cutoff date (November 20, 2023). Patients who signed the informed consent form and completed at least the eCRF were eligible for baseline analysis.

Ethics approval for this study was granted by the Ethics Committee of the North Rhine Medical Association (Ärztekammer Nordrhein) on October 13, 2022 (approval number: 2022263). All patients were offered the informed consent form; of the 103 who signed the form, 2 withdrew consent. All patient data were handled in a pseudonymized manner, and the privacy of patients was ensured. Upon completion of the baseline eCRF (regardless of EQ-5D-5L or DTSQ completion), patients received a €10 (US $10.80) voucher for Shop Apotheke as compensation for their time spent on data documentation.

Data were analyzed descriptively with SAS (release 9.4 or higher; SAS Institute Inc). UACRs were categorized based on KDIGO ACR stages [6], with the addition of a fourth level indicating nephrotic-range albuminuria [28,29] (stage A1: UACR of <30 mg/g; stage A2: UACR of 30‐300 mg/g; stage A3: UACR of 301‐3000 mg/g; stage A4: UACR of >3000 mg/g). Serum fasting glucose levels were categorized based on the ADA diagnostic criteria for diabetes and prediabetes [30], and eGFRs were reported by KDIGO category [6]. The EQ-5D-5L index score was calculated by using a value set for the German general population [31].

During the awareness campaign, push notifications for study- and indication-related content were sent out to a cohort of patients with T2DM (Figure 1), whereof 8799 to 9527 patients received these push notifications. Of these 8799 to 9527 patients, 103 provided consent, and 2 withdrew consent. A total of 101 patients completed the eCRF and were therefore included in the baseline analysis.

Of the 101 participating patients with T2DM, 61 were male, and 40 were female. The study population had a mean age of 54.2 (SD 11.4; range 22‐76) years, a mean weight of 103.9 (SD 21.7; range 45‐149) kg, and a mean BMI of 34 (SD 7.0; range 16.9‐54.7) kg/m². Further, 26 patients were smokers, 43 were former smokers, and 32 had never smoked.

Of the 101 patients in the analyzed cohort, only 1 female patient and 5 male patients reported their UACR values (Table 1); 3 of the patients (all male) had a UACR of ≥30 mg/g. The remaining 95 (94.1%) patients reported that their HCPs had not provided UACR measurements. For eGFR values, a similar finding was observed—only 9 (8.9%) of the 101 patients were aware of their latest values.

In contrast to laboratory values provided by HCPs, self-determined test results were more commonly reported. In total, 65 (64.4%) patients provided their serum fasting glucose levels, and 90 (89.1%) patients provided their latest HbA_1c values (Table 2).

Systolic and diastolic blood pressures are presented in Table 3. In total, 31 (30.7%) patients had a high-normal systolic blood pressure, 15 (14.9%) patients showed systolic hypertension, and 13 (12.9%) patients showed diastolic hypertension according to the 2023 European Society of Hypertension guidelines for the management of arterial hypertension (assuming office-based blood pressure measurements) [32].

Use of SGLT2is, the nonsteroidal mineralocorticoid receptor antagonist finerenone, hypertension medication (defined by anatomic therapeutic chemical codes), and nephrotoxic medication (as listed by Patel and Sapra [33]) was reported by 44, 0, 80, and 83 patients, respectively. All patients confirmed their T2DM diagnosis at baseline. Further, 81 reported hypertension (77 with treated hypertension and 4 with untreated hypertension) and 12 reported CKD as comorbidities. Reported comedication data for all patients are presented in Multimedia Appendix 1.

A total of 50 patients completed the EQ-5D-5L. The mean EQ-visual analog scale score was 64.4 (SD 24.4; range 3.0‐99.0; on a scale of 0‐100, where 0 is the worst and 100 is the best health state possible), and the mean EQ-5D-5L index score was 0.7 (SD 0.3; range –0.1 to 1.0). The DTSQ was completed by 49 patients, with a mean DTSQ score of 28.8 (SD 6.8; range 9.0‐36.0) at baseline.

With the SMART-Finder study, we are investigating the self-reporting of UACR, eGFR, and HbA_1c values in a cohort of patients with T2DM using an adherence-supporting app in Germany. This baseline analysis provides insights into the implementation of treatment guidelines and HCP-patient interactions.

We observed a low participation rate in our cohort of patients, who were taking responsibility for their health status by regularly tracking adherence. Of the 9527 patients who were contacted, only 712 (7.5%) to 1659 (17.4%) patients engaged with at least one aspect of the awareness campaign by clicking on a video or the written information content in the app. The final participation rate—8.4% (101 of 1209 patients who saw the SMART-Finder study video)—is comparable to those observed in other awareness campaigns; for example, 10.2% (804/7865) and 10.1% (802/7920) of invited participants completed a survey on a public health campaign for increasing kidney health awareness in a Canadian province [34].

In contrast to activities that can be self-managed by patients (blood pressure measurement and testing of HbA_1c and serum fasting glucose levels), about which our cohort appears to be well informed, our results suggest a dramatic lack of awareness of renal function status in this population at risk of CKD. This is of particular concern, since almost half of the cohort (46/101, 45.5%) had a systolic blood pressure of ≥130 mm Hg, and 82.2% (83/101) reported former or current intake of nephrotoxic medication. Further, more than 90% of the patients were unaware of their eGFR (92/101, 91.1%) and UACR (95/101, 94.1%), but 89.1% (90/101) of the patients knew their HbA_1c level, suggesting that treating physicians either do not test for UACRs and eGFRs or do not inform patients about the results. Both a lack of adherence to guidelines for diagnostic testing and a lack of HCP-patient communication put patients at risk. Another reason for this imbalance in health literacy may be the focus on HbA_1c, rather than on kidney comorbidity, in patient education material.

The presented baseline analysis has some limitations. There is a selection bias, as only users of the adherence-supporting app, that is, those who seemed to have experience with using technology, were able to participate. Additionally, the unconventional method of data collection, with patients requesting data from HCPs, may have introduced reporting bias from the HCPs. Further, due to the low participation rate and the lack of UACR and eGFR values, our findings on CKD risk in the observed cohort of patients may not be generalizable to all patients with T2DM. However, the missing data show a lack of awareness in a high-risk population. Even though the annual eGFR and albuminuria testing rates in Germany are 96.5% and 77.2%, respectively, our results suggest that among German patients with diabetes, knowledge regarding their kidney function and disease is still insufficient [35].

Baseline data from the SMART-Finder study, which were derived from patients with T2DM who were using an adherence-supporting app in Germany, showed a dramatic lack of awareness of kidney-related biomarkers (UACR and eGFR) but revealed high knowledge of HbA_1c values. As only informed patients are empowered patients, future goals for diabetes management must include guideline-compliant testing of CKD-related biomarkers and open communication between HCPs and patients.