Abstract
We evaluated the effectiveness of a mobile health (mHealth) intervention for diabetic kidney disease patients by conducting a 12-month randomized controlled trial among 126 type 2 diabetes mellitus patients with moderately increased albuminuria (urinary albumin-to-creatinine ratio (UACR): 30-299 mg/g creatinine) recruited from eight clinical sites in Japan. Using a Theory of Planned Behavior (TPB) behavior change theory framework, the intervention provides patients detailed information in order to improve patient control over exercise and dietary behaviors. In addition to standard care, the intervention group received DialBetesPlus, a self-management support system allowing patients to monitor exercise, blood glucose, diet, blood pressure, and body weight via a smartphone application. The primary outcome, change in UACR after 12 months (used as a surrogate measure of renal function), was 28.8% better than the control group’s change (P = 0.029). Secondary outcomes also improved in the intervention group, including a 0.32-point better change in HbA1c percentage (P = 0.041). These improvements persisted when models were adjusted to account for the impacts of coadministration of drugs targeting albuminuria (GLP-1 receptor agonists, SGLT-2 inhibitors, ACE inhibitors, and ARBs) (UACR: −32.3% [95% CI: −49.2%, −9.8%] between-group difference in change, P = 0.008). Exploratory multivariate regression analysis suggests that the improvements were primarily due to levels of exercise. This is the first trial to show that a lifestyle intervention via mHealth achieved a clinically-significant improvement in moderately increased albuminuria.
Similar content being viewed by others
Introduction
Globally, 537 million people aged 20–79 suffer from diabetes1. Diabetic kidney disease (DKD), one of the most common and costly complications of diabetes2, is a leading cause of kidney failure and significantly increases cardiovascular disease3,4. Urinary albumin-to-creatinine ratio (UACR) is an independent factor in predicting kidney prognosis, diabetic retinopathy, and macrovascular disorders. Reduction of albuminuria is associated with decreased risk of death and kidney failure5,6,7, with a 30% one-year reduction associated with improved cardiovascular and kidney outcomes8. UACR correlates well with renal outcome and is an important surrogate endpoint in clinical trials5,6,9.
Moderately increased albuminuria is associated with cardiovascular events, and controlling it requires a multifactorial approach. Mobile Health (mHealth, the use of mobile phones and other connected devices to improve health) can provide such an approach. Although pharmacological interventions such as renin-angiotensin-aldosterone system (RAAS) inhibitors, glucagon-like peptide (GLP)-1 receptor agonists, and sodium-glucose cotransporter (SGLT)-2 inhibitors can slow DKD’s progression, polypharmacy and adverse drug events pose clinical challenges10. Lifestyle modifications for diet and physical activity along with improved management of glucose, blood pressure (BP), and lipids are a cost-effective therapeutic adjunct to pharmacological treatment11,12,13,14. Improvements in HbA1c are associated with improved renal outcomes15,16. There is evidence that lifestyle modifications can improve kidney function and postpone the progression of DKD17,18,19, though the mechanisms are unclear and research in the area is ongoing. Interventions using mHealth to support patient lifestyle self-management have been shown to control blood glucose levels in type 2 diabetes (T2D) patients20,21 and improve physical activity, diet, and medication adherence22, but have not previously been shown to improve kidney function. Our mHealth intervention, DialBetesPlus, is an improved version of DialBetics, a self-management support system for patients with T2D that improved HbA1c by 0.4% in a three month 54-patient randomized controlled trial (RCT)23. DialBetesPlus helps patients monitor exercise, blood glucose, diet, blood pressure, and body weight via a smartphone application.
This study adds DialBetesPlus to usual care among patients with early-stage DKD to assess its impact, relative to a control group receiving usual care, on albuminuria, kidney function, glycemic control, BP, lipid profile, body mass index (BMI), quality of life (QOL), and diabetes self-management.
Results
We started recruiting patients on July 1, 2018, and we completed recruitment on August 31, 2019, after enrolling 159 patients. We ended the study on April 6, 2021 upon completion of the follow-up at week 72 for the final patient. 27 of 159 subjects were ineligible—nine identified as such before the two week device trial, and 18 identified after the trial (two newly recognized as not meeting inclusion criteria, seven not able or willing to use the device for at least seven days, and nine withdrawing consent for various personal reasons). The remaining 132 subjects were randomly assigned between groups, yielding 66 (intervention) and 60 (control) at baseline and 62 (intervention) and 60 (control) at 12 months, with 59 (control) having paired baseline and 12-month UACR and HbA1c values (Fig. 1). All 66 intervention and 60 control group patients were included in our Full Analysis Set results, including three intervention patients who were non-compliant to the protocol. Baseline characteristics were well matched (Table 1), although a significantly higher proportion in the intervention group were taking the GLP-1 receptor agonists (37.9% versus 15.0% in the control group, P = 0.005).
Effect of intervention on behavior
The intervention uses monitoring and feedback with a goal of increasing exercise behavior. The mean monthly exercise measurement rate was high (68.5–81.6%, Fig. 2a). SDSCA results were good, with significant improvement relative to the control group (6-month, P = 0.029; 18-month, P = 0.022; Fig. 2d, g), indicating improvements in lifestyle compliance. Daily step counts (monthly means of 7552–8693, Fig. 3) stayed consistent throughout the intervention period. Although measurement rates decreased over the course of the intervention, falling from 80.6% to 68.5%, the SDSCA results show that the intervention increased exercise levels, with the effect persisting after the intervention (at month 18).
Similarly, the intervention uses monitoring and feedback to increase blood glucose monitoring, with results that also show a significant improvement relative to the control group (6-month, P < 0.001; 12-month, P < 0.001). Measurement rates were high, with monthly means falling from 73.4% to 59.0% over the intervention. SDSCA results show improvement (Fig. 2b, e, h), though the improvement disappeared after the intervention (at month 18). Finally, for diet, monthly mean meal measurement rates were low and declined from 54.0% to 37.2% over the intervention, and the intervention did not show significant improvement relative to the control group (Fig. 2c, f, i).
The intervention group’s monthly mean study retention rate at 12 months was very high, at 93.9% (62/66), while monthly mean engagement rate with DialBetesPlus was high, at 81.6% at 12 months. Average daily measurement rate of body weight fell from 77.5% to 61.7%, with measurement of morning BP falling from 71.5% to 59.1%. There were no dropouts in the intervention group due to DialBetesPlus systems failures.
Effect of intervention on health
The intervention group saw a range of changes in UACR, with most patients improving (Fig. 4a). The primary outcome, change in UACR after 12 months, had a statistically and clinically significant between-group difference in change of −28.8% (P = 0.029) (Table 2). The intervention group had nearly twice as many patients with a reduction of ≥30% of baseline24 UACR (25/62 (40.3%) vs. 12/59 (20.3%), relative risk = 1.98, P = 0.019.) For secondary outcomes, HbA1c had a similar range of changes (Fig. 4b), with a statistically and clinically significant between-group difference in change of −0.32 points (P = 0.041) (Table 2). HDL-C had a statistically and clinically significant improvement compared to the control group (P = 0.041). While the intervention group had better eGFR (between-arm difference of −2.3 mL/min/1.73m2), suggesting better kidney function, the difference did not reach statistical significance (P = 0.141). Other parameters showed no significant difference between the two groups. Improvements in UACR and HbA1c continued at the month 18 follow-up, although improvements in the control group made the difference no longer significant.
To exclude the possible influence of baseline differences between the control and intervention groups in coadministration of drugs known to improve albuminuria, we performed a post hoc assessment of outcomes via analysis of covariance (ANCOVA) using as covariates the baseline value of the outcome under analysis along with the use of each of GLP-1 receptor agonists, SGLT-2 inhibitors, ACE inhibitors, and ARBs at baseline, along with a subgroup analysis of UACR eliminating 14 patients (ten intervention, four control) whose use of these albuminuria-related drugs intensified during the first six months of the study. The results (Supplementary Table 1, Supplementary Figure 1) show improvements that are very similar to those in Table 2, with similar improvements in BMI now reaching statistical significance (P = 0.045). Removing outliers from the analysis as a sensitivity test also produced similar results.
Our post hoc exploratory multivariate linear regression analysis of change in log-UACR (Supplementary Table 2) showed statistical significance for the resulting model’s variables of baseline value of log-UACR (P < 0.001), the change in HbA1c (P = 0.003), and the change in sBP (P = 0.010). There was no statistical significance for the variable of change in blood glucose. When using the mean change in inputs, only the change in HbA1c achieved a clinically significant model impact, with a −10.8% change in UACR from a change in HbA1c of −0.28 points.
Since the multivariate regression analysis of log-UACR suggests that change in HbA1c was a key driver, we did a similar post hoc multivariate regression analysis of change in HbA1c (Supplementary Table 3). The resulting model shows statistical significance for all four variables: baseline value of HbA1c (P < 0.001), daily steps (P = 0.045), change in blood glucose (P = 0.002), and change in BMI (P = 0.011). Using the mean values for the inputs results in clinically significant (−0.35 point change from 7560 steps) and weaker (−0.15 point change from −0.7 kg/m2 change in BMI, 0.07 point change from 4.07 mm/dL change in glucose) changes in modeled HbA1c percentage.
These regression analyses, while not conclusive, suggest a primary biological mechanism for the results:
-
(1)
The intervention improved HbA1c via improvements in exercise.
-
(2)
The physiological changes reflected in this improvement in HbA1c in turn led to an improvement in UACR.
There were no all-cause mortality events, composite cardiovascular outcomes, composite kidney endpoints, severe hypoglycemic events, or adverse events. We found no significant between-arm changes in ADDQoL.
Discussion
In this study, we employed albuminuria as a surrogate endpoint for predicting kidney and cardiovascular outcomes. Evidence suggests that albuminuria is closely associated with significant clinical endpoints and acts as an early indicator of disease progression, making it a valid and widely used surrogate5,6,9. However, surrogate endpoints have inherent limitations, including their inability to directly measure clinical outcomes like ESKD, creatinine doubling, and cardiovascular death. Consequently, future research is necessary to determine whether lifestyle interventions via mHealth can genuinely impact these hard endpoints beneficially.
Studies have shown association between physical activity and improvements in UACR and kidney function, including reduction to the risk of renal failure25. Exercise-based lifestyle interventions have shown a causal link to improved UACR26,27,28,29. To our knowledge, the current study is the first to show significant improvements in UACR from an mHealth-based lifestyle intervention.
The relative UACR change of −28.8% (−32.3% using ANCOVA) in the present study is comparable in magnitude to 17–32%30,31, 2–39%32, and 29%33 changes from reported drug-based albuminuria-targeted interventions, and this magnitude of change seems to be clinically relevant. However, the mechanism underlying the improvement in UACR seen in this intervention may differ from that of drug-based interventions, and we cannot definitively conclude that the intervention improved long-term kidney health. We are unaware of any studies linking a lifestyle intervention’s improvement in UACR to long-term kidney health, and this is an area that needs further work.
Improved HbA1c has a proven impact on kidney health, and the study’s relative HbA1c change of −0.32 points corresponds to approximately one additional year of sight, absence of kidney failure, absence of lower extremity amputation, and life34. The findings suggest that the mHealth intervention employing DialBetesPlus was effective in delaying the progress of DKD.
Our somewhat elderly participants could use the system without burden. Our high engagements are promising, as initial engagement with mHealth applications is closely related to long-term engagement35. Automated pedometer counting had the highest engagement, though measurement rates declined with time. A previous multicomponent mHealth trial showed similar decreases with time36. Personalization of mHealth content, social and gamification features, and personal support have been shown to improve adherence to mHealth37. There is a need to further explore mechanisms to improve long-term engagement.
The mHealth intervention by DialBetesPlus seems to have led to increased exercise, as measured via the SDSCA exercise score, and may have improved insulin sensitivity, reducing cardiovascular risk and improving albuminuria38,39,40. We do not have a baseline step count, but the high step count, 7000–8000 (7552 ± 4449) steps per day, corresponds well with the recommended 150 min per week of activity to prevent all-cause mortality41,42,43,44 and results showing that increasing to 6000–8000 daily steps decreased mortality among adults aged 60 years and older. Blood glucose monitoring had good engagement but did not definitively lead to improved outcomes in this study. The diet feature had low engagement and SDSCA scores indistinguishable from those of the control group, and we conclude that this feature was not effective as currently implemented. In terms of the TPB framework, the intervention did not succeed in improving dietary control, presumably because, relative to the comparatively simple choice of whether or not to exercise, diet involves complicated choices throughout the day balancing practicalities (for instance, the availability of prepared food), desirabilities (taste, satisfying hunger, etc.) and multiple dimensions of nutrition (calories, macronutrients, fiber, salt, etc.). There seems to be a need for more specific and individualized information to support patient control over dietary behavior. Future work to improve the engagement with and effectiveness of features addressing meal monitoring and assess any resulting health improvements is warranted.
We did not find significant improvement in BP, despite increased exercise, perhaps because of limited room for improvement given fairly well-controlled baseline BP.
We are unable to definitively evaluate which features contributed most to improved outcomes, although we expect improvements are related to the features with the highest engagement. Features could have interacted with one another. The addition of a self-management support system could have enhanced patient engagement, a critical component of successful treatment of T2D and DKD45,46, and may have improved patient focus on multifactorial intervention47.
The intervention is general in nature and may be readily extended to other patient populations. The Japanese guidelines for lifestyle modifications are similar to those in other countries, and we expect this intervention would show similar positive results, including improved albuminuria, for diabetes patients in other countries. More broadly, the intervention could apply for any patient population who would benefit from these lifestyle modifications, not just diabetes patients. At present, the DialBetesPlus application is a research product, not production software, and applying this intervention more broadly, whether for diabetes patients in Japan or for a wider population, would likely require an industry partner who would develop a full production application.
Use and changes in use of RAAS inhibitors, GLP-1 receptor agonists, and SGLT-2 inhibitors were allowed during the study, potentially confounding any effect of the intervention on UACR. Our analysis adjusting for usage prior to the intervention and for intensification during the intervention showed similar UACR results, reducing the chance that these agents partially account for our results. Similarly, our exploratory multivariate regression analysis did not show that intensification of diabetes drugs was a significant driver of change in HbA1c. Our lack of restriction on concurrent diabetes management allowed the study to be conducted in a condition close to that of the real world.
The study has limitations. The study was conducted among a Japanese population, and there are differences between Japanese and other populations in lifestyles and in the pathophysiology of T2D and DKD. The study was limited to patients able to use mobile phones, so there may be biases due to users’ digital literacy. Participants were not blinded to randomization, so social desirability bias may have affected the results. The study’s focus on patients with mild DKD may limit generalizability to those with more advanced DKD. Albuminuria is only one aspect of DKD—further research on the intervention’s impact on broader aspects of kidney function is warranted. Our use of single UACR measures at each time point, first-morning void as a good alternative to measuring 24-h urinary albumin excretion48, may cause a bias toward the null. Many factors influence UACR, not all of which were captured in these analyses. For example, we were unable to capture the effects of changes in albuminuria-related drugs between six and 12 months, and our measurements of diet were ineffective, a clear weakness of our study. As clinical studies of DKD with digital health are relatively new, we were unable to use previous studies to fully plan statistical analyses or develop hypotheses, and therefore some of the analyses were post hoc and exploratory. These results should be considered hypotheses.
This study is, to our knowledge, the first to show that the addition of a real-time self-management mHealth system to standard care reduced albuminuria. Glycemic control improved and was comparable to that of other mHealth interventions incorporating coaching from health care providers49,50. This intervention does not require healthcare personnel, so it has great potential for scalability. As the use of smart phones and the mobile internet becomes ubiquitous in daily life, a self-management support mHealth system is recommended to reduce the lifestyle risk factors of DKD patients. Future studies are needed to improve long-term engagement with the mHealth system and to develop effective implementations within the broader healthcare system.
Methods
Participant inclusion criteria and recruitment
Inclusion criteria were outpatients to the eight registered medical institutions who were diagnosed with type 2 diabetes with moderately increased albuminuria (UACR: 30–299 mg/g creatinine). The institutions (Supplementary Table 4) as well as inclusion and exclusion criteria (Supplementary Table 5) were described in detail previously51. Medication therapy (antidiabetic, lipid lowering, and antihypertensive) could be adjusted as needed during the study consistent with standard care.
After asking all enrolled patients to use DialBetesPlus for two weeks, we retained as eligible for the study those who used the application and devices for at least seven days during the two weeks. (We judged that the two-week trial would have no significant effect on outcomes.) We then used SAS 9.4 to randomize eligible participants one-to-one to either the intervention or the control group using the covariance-adaptive randomization by minimization method with random element of 0.75 to ensure covariance balance for age (≤40, 41–59, ≥60), sex, and UACR (<100, ≥100), stratified by institution52. Our protocol defined participants to be dropouts if there was no data input after three weeks or if the research team lost contact with the participants.
DialBetesPlus design
The DialBetesPlus intervention seeks to change patient behavior by improving adherence to existing prescribed lifestyle modifications, rather than introducing a new treatment. The desired behavior changes are increased exercise and a move to diabetes-appropriate dietary choices, and these behavior changes can be viewed using the framework of the Theory of Planned Behavior (TPB). TPB is a behavior change theory that has seen wide use and success in physical behavior interventions53. At a high level, the TPB framework models behavior as driven by the individual’s intention to perform the behavior and by their control (both perceived and actual) over the behavior. Using the TPB framework, the DialBetesPlus intervention does not target intention, based on an assumption that most patients are well motivated to address this serious disease. Rather, the intervention seeks to strengthen both perceived and actual control by giving patients timely and detailed information on behavior (exercise and diet) as well as intermediate health outcomes (blood glucose, BP, and body weight). This feedback allows patients to see which individual behavioral decisions lead to success in achieving the desired behavior and intermediate health outcomes, with the goal of ultimately improving glycemic control and slowing the progression of kidney disease.
Using DialBetesPlus (Fig. 5)51, patients measure daily step counts, blood glucose, BP, and body weight at home. Data is transferred from each device to the DialBetesPlus smartphone application and immediately sent to the DialBetesPlus server and evaluated following the target values of the Japan Diabetes Society (JDS) guidelines (http://www.fa.kyorin.co.jp/jds/uploads/Treatment_Guide_for_Diabetes_2016-2017.pdf, accessed on April 18, 2023). The step target is 8000 or more steps per day, and target maximum values are blood glucose levels of 110 mg/dl before breakfast and 140 mg/dl at bedtime, and BP of 125/75 mmHg. Feedback is sent to the participant’s smartphone (Fig. 6). In addition, patients enter the content and quantity of their meals by text message with a photograph of the meal. Software on the server calculates nutrient intake and consumed calories, generates JDS-based advice for improving dietary habits, and sends the measurements and advice to the smartphone. Patients are also able to enter the type and duration of exercise that was not recorded on the pedometer. Patients can review their data in the forms of graphs.
Study design
The study was a prospective, randomized, open-label, multicenter clinical trial. M.M. generated the random allocating sequence used to randomize the participants into either the intervention or control group in a one-to-one fashion based on albuminuria levels, gender, and age. After randomization, the principal investigator (K.W.) communicated the assignment to the study collaborators (Y.K., K.M., Y.T., S.K., A.I., S.S., T.M., and T.T.) who enrolled the participants. Patients assigned to the intervention group used DialBetesPlus for 12 months, with a post-intervention follow-up at month 18. We provided the intervention group patients with an NFC-enabled blood glucose meter (MS-FR201B; Terumo), a Bluetooth-enabled BP monitor (HEM-7271T; Omron), an NFC-enabled pedometer (MT-KT02DZ; Terumo), and a Bluetooth-enabled scale (HBF-255T; Omron). The devices were paired with a smartphone (Arrows F-02H: Fujitsu or Galaxy Note3 SC-01F: Samsung) prepared with the DialBetesPlus application. We provided participants in the control group with a home BP monitor as part of standard care. We established each patient’s baseline by collecting background demographic information, physical parameters (BMI and BP), blood test data, UACR by first morning void, medication regimen, and patients’ QOL and self-care evaluation via questionnaire, with follow-up studies at two, six, 12 and 18 months. (UACR was obtained only at baseline, month 12, and month 18.).
Outcome measures
The primary outcome was change after 12 months in UACR. Secondary outcomes (Supplementary Table 6) included change after 12 months in blood test parameters, physical parameters (BMI and BP), lifestyle habits and diabetes self-care assessment, and QOL, along with changes in medication therapy, study retention rate, engagement rate, and daily measurement rate of DialBetesPlus. We tracked any DialBetesPlus system failures. To ensure safety, we monitored the number of hypoglycemic events and other adverse events.
Research ethics
We conducted this study in compliance with the Helsinki Declaration. The Research Ethics Committee of the Graduate School of Medicine, the University of Tokyo and related facilities approved the study protocol and informed consent form. We registered the study in the University Hospital Medical Information Network Clinical Trial Register (UMIN000033261). We obtained written informed consent from all participants prior to the study.
Statistical analysis
We calculated the required sample size of 64, based on our target of 80% power at a two-sided significance level of 0.05 and assuming a between-arm difference of change in UACR of 100 mg/gCre with a standard deviation (SD) of 20054,55. Assuming a 20% loss rate, we required 80 initial subjects per group.
The study uses the Full Analysis Set of all patients after randomization for which data were obtained at least once. Data on patients’ characteristics are presented as mean and SD for continuous variables that follow a normal distribution, and median and interquartile ranges for continuous variables with a non-normal distribution. Categorical variables are presented as frequency and proportion. We compared continuous variables using a two-sided t-test with a significance level of 5% or a Wilcoxon rank sum test. For UACR, we used logarithmically transformed data (log-UACR) due to a skewed distribution. As the levels of missing data were low, with 4.0% (5/126) of patients missing UACR data and 4.0% missing HbA1c data, we used standard methods56,57 without imputation for missing data. A secondary analysis using imputation (multiple imputation by chained equations procedure and 100 imputations for missing data) showed similar results (results not shown.) We analyzed categorical variables using Fisher’s exact test or the Chi square test, changes in J-SDSCA and JP-ADDQoL scores using t-test, and changes in medication regimen at 12 months using Fisher’s exact test.
We performed post hoc sensitivity analyses to assess the effect of clinically relevant covariates on outcomes and explore possible mechanisms underlying our results. To exclude the possible influence of baseline differences between the groups in coadministration of drugs known to improve albuminuria, we assessed outcomes via ANCOVA using as covariates the baseline value of the outcome under analysis along with the use at baseline of each of GLP-1 receptor agonists, SGLT-2 inhibitors, Angiotensin-converting enzyme (ACE) inhibitors, and Angiotensin receptor blockers (ARBs). We also performed a subgroup analysis of the ANCOVA results for UACR wherein we eliminated patients whose use of these albuminuria-related drugs intensified during the first six months of the study.
To address our secondary objective of identifying physical/biological parameters most strongly related to UACR and HbA1c reduction, we conducted exploratory multivariate linear regressions. Based on previous research and clinical judgment, we included candidate variables with established or presumed importance in UACR (13 variables) and HbA1c (13 variables) (Supplementary Table 7). We used the best subset method, fitting separate regression models for all combinations of up to four variables, based on 66 patients and the criterion of at least 15 patients per variable58, to determine which model is best59,60 based on the criteria of highest adjusted R2.
We defined statistical significance as a two-sided P-value less than 0.05 and used SAS (version 9.4 M7; SAS Institute Inc) for statistical analysis.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request for non-commercial purposes.
Code availability
The code used for the analysis in this study is based on a common procedure used in the statistical analysis software SAS. No mathematical algorithms or/and original code are used.
Change history
15 June 2024
A Correction to this paper has been published: https://doi.org/10.1038/s41746-024-01158-w
References
Magliano D. J., Boyko E. J. IDF Diabetes Atlas 10th edition scientific committee. IDF DIABETES ATLAS [Internet]. 10th ed. Brussels: International Diabetes Federation; 2021.
Gülümsek, E. Keşkek ŞÖ. Direct medical cost of nephropathy in patients with type 2 diabetes. Int Urol. Nephrol. 54, 1383–1389 (2022).
Koye, D. N. et al. The global epidemiology of diabetes and kidney disease. Adv. Chronic Kidney Dis. 25, 121–132 (2018).
Tofte, N. et al. Early detection of diabetic kidney disease by urinary proteomics and subsequent intervention with spironolactone to delay progression (PRIORITY): a prospective observational study and embedded randomised placebo-controlled trial. Lancet Diabetes Endocrinol. 8, 301–312 (2020).
Coresh, J. et al. Change in albuminuria and subsequent risk of end-stage kidney disease: an individual participant-level consortium meta-analysis of observational studies. Lancet Diabetes Endocrinol. 7, 115–127 (2019).
Heerspink, H. J. L. et al. Change in albuminuria as a surrogate endpoint for progression of kidney disease: a meta-analysis of treatment effects in randomised clinical trials. Lancet Diabetes Endocrinol. 7, 128–139 (2019).
Penno, G. et al. Independent correlates of urinary albumin excretion within the normoalbuminuric range in patients with type 2 diabetes: The Renal Insufficiency And Cardiovascular Events (RIACE) Italian Multicentre Study. Acta Diabetol. 52, 971–981 (2015).
Persson, F. et al. Changes in Albuminuria Predict Cardiovascular and Renal Outcomes in Type 2 Diabetes: A Post Hoc Analysis of the LEADER Trial. Diabetes Care 44, 1020–1026 (2021).
Levey, A. S. et al. Change in Albuminuria and GFR as End Points for Clinical Trials in Early Stages of CKD: A Scientific Workshop Sponsored by the National Kidney Foundation in Collaboration With the US Food and Drug Administration and European Medicines Agency. Am. J. Kidney Dis. 75, 84–104 (2020).
Mendonça, L. et al. Characterizing palliative care needs in people with or at risk of developing diabetic foot ulcers. Ther. Adv. Endocrinol. Metab. 13, 20420188221136770 (2022).
Stevens, P. E. & Levin, A. Evaluation and management of chronic kidney disease: synopsis of the kidney disease: improving global outcomes 2012 clinical practice guideline. Ann. Intern. Med. 158, 825–830 (2013).
Eckardt, K. U. et al. Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 93, 1281–1292 (2018).
Guideline development group. Clinical Practice Guideline on management of patients with diabetes and chronic kidney disease stage 3b or higher (eGFR <45 mL/min). Nephrol. Dial. Transpl. 30, ii1–ii142 (2015).
Onyenwenyi, C. & Ricardo, A. C. Impact of lifestyle modification on diabetic kidney disease. Curr. Diab Rep. 15, 60 (2015).
Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of Diabetes on the development and progression of long-term Complications in insulin-dependent Diabetes Mellitus. N. Engl. J. Med. 329, 977–986 (1993).
Stratton, I. M. et al. Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ 321, 405–412 (2000).
Amaral, L. S. B., Souza, C. S., Lima, H. N. & Soares, T. J. Influence of exercise training on diabetic kidney disease: a brief physiological approach. Exp. Biol. Med. 245, 1142–1154 (2020).
Sheshadri, A. et al. Structured moderate exercise and biomarkers of kidney health in sedentary older adults: the lifestyle interventions and independence for elders randomized clinical trial. Kidney Med. 5, 100721 (2023).
Dong, L. et al. Long-term intensive lifestyle intervention promotes improvement of stage III diabetic nephropathy. Med Sci. Monit. 25, 3061–3068 (2019).
Cui, M. et al. T2DM self-management via smartphone applications: a systematic review and meta-analysis. PLoS ONE 11, e0166718 (2016).
Greenwood, D. A. et al. A systematic review of reviews evaluating technology-enabled diabetes self-management education and support. J. Diabetes Sci. Technol. 11, 1015–1027 (2017).
Changizi, M. & Kaveh, M. H. Effectiveness of the mHealth technology in improvement of healthy behaviors in an elderly population-a systematic review. Mhealth 3, 51 (2017).
Waki, K. et al. DialBetics: A Novel Smartphone-based Self-management Support System for Type 2 Diabetes Patients. J. Diabetes Sci. Technol. 8, 209–215 (2014).
Bhalla, V. et al. Racial/ethnic differences in the prevalence of proteinuric and nonproteinuric diabetic kidney disease. Diabetes Care 36, 1215–1221 (2013).
Cai, Z., Yang, Y. & Zhang, J. Effects of physical activity on the progression of diabetic nephropathy: a meta-analysis. Biosci. Rep. 41, BSR20203624 (2021).
Lazarevic, G. et al. Effects of aerobic exercise on microalbuminuria and enzymuria in type 2 diabetic patients. Ren. Fail 29, 199–205 (2007).
Hellberg, M. et al. Randomized controlled trial of exercise in CKD-The RENEXC study. Kidney Int. Rep. 4, 963–976 (2019).
Sokolovska, J. et al. Impact of interval walking training managed through smart mobile devices on albuminuria and leptin/adiponectin ratio in patients with type 2 diabetes. Physiol. Rep. 8, e14506 (2020).
Yamamoto-Kabasawa, K. et al. Benefits of a 12-week lifestyle modification program including diet and combined aerobic and resistance exercise on albuminuria in diabetic and non-diabetic Japanese populations. Clin. Exp. Nephrol. 19, 1079–1089 (2015).
Mann, J. F. E. et al. Liraglutide and renal outcomes in Type 2 diabetes. N. Engl. J. Med. 377, 839–848 (2017).
von Scholten, B. J. et al. The effect of liraglutide on renal function: a randomized clinical trial. Diabetes Obes. Metab. 19, 239–247 (2017).
Muskiet, M. H. A. et al. Lixisenatide and renal outcomes in patients with type 2 diabetes and acute coronary syndrome: an exploratory analysis of the ELIXA randomised, placebo-controlled trial. Lancet Diabetes Endocrinol. 6, 859–869 (2018).
Tuttle, K. R. et al. Dulaglutide versus insulin glargine in patients with type 2 diabetes and moderate-to-severe chronic kidney disease (AWARD-7): a multicentre, open-label, randomised trial. Lancet Diabetes Endocrinol. 6, 605–617 (2018).
Chetty, V. T. et al. The effect of continuous subcutaneous glucose monitoring (CGMS) versus intermittent whole blood finger-stick glucose monitoring (SBGM) on hemoglobin A1c (HBA1c) levels in Type I diabetic patients: a systematic review. Diabetes Res Clin. Pr. 81, 79–87 (2008).
Spring, B. et al. Multicomponent mHealth intervention for large, sustained change in multiple diet and activity risk behaviors: the make better choices 2 randomized controlled trial. J. Med. Internet Res. 20, e10528 (2018).
Chawla, N. V. & Davis, D. A. Bringing big data to personalized healthcare: a patient-centered framework. J. Gen. Intern. Med. 28, S660–S665 (2013).
Jakob, R. et al. Factors influencing adherence to mHealth apps for prevention or management of noncommunicable diseases: systematic review. J. Med. Internet Res. 24, e35371 (2022).
Shaw, K. A., Gennat, H. C., O’Rourke, P. eds. Exercise for overweight or obesity. Cochrane Database of Systematic Reviews, (4), John Wiley & Sons, Ltd; 2006.
Câmara, N. O. et al. Kidney disease and obesity: epidemiology, mechanisms and treatment. Nat. Rev. Nephrol. 13, 181–190 (2017).
Navaneethan, S. D. et al. Urinary albumin excretion, HMW adiponectin, and insulin sensitivity in type 2 diabetic patients undergoing bariatric surgery. Obes. Surg. 20, 308–315 (2010).
Tudor-Locke, C. et al. Accelerometer steps/day translation of moderate-to-vigorous activity. Prev. Med. 53, 31–33 (2011).
Del Pozo Cruz, B. et al. How many steps a day to reduce the risk of all-cause mortality? A dose-response meta-analysis. J. Intern. Med. 291, 519–521 (2022).
Lee, I. M. et al. Association of step volume and intensity with all-cause mortality in older women. JAMA Intern. Med, 179, 1105–1112 (2019).
Bull, F. C. et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br. J. Sports Med. 54, 1451–1462 (2020).
Rossing, P. & Epstein, M. Microalbuminuria constitutes a clinical action item for clinicians in 2021. Am. J. Med. 135, 576–580 (2022).
Shlipak, M. G. et al. The case for early identification and intervention of chronic kidney disease: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 99, 34–47 (2021).
Minutolo, R. et al. Cardiorenal prognosis by residual proteinuria level in diabetic chronic kidney disease: pooled analysis of four cohort studies. Nephrol. Dial. Transpl. 33, 1942–1949 (2018).
Lambers Heerspink, H. J. et al. Albuminuria assessed from first-morning-void urine samples versus 24-hour urine collections as a predictor of cardiovascular morbidity and mortality. Am. J. Epidemiol. 168, 897–905 (2008).
Ang, I. Y. H. et al. A Personalized Mobile Health Program for Type 2 Diabetes During the COVID-19 Pandemic: Single-Group Pre–Post Study. JMIR Diabetes 6, e25820 (2021).
Riangkam, C. et al. Effects of a mobile health diabetes self-management program on HbA1C, self-management and patient satisfaction in adults with uncontrolled type 2 diabetes: a randomized controlled trial. J. Health Res. 36, 878–888 (2022).
Kawai, Y. et al. Efficacy of the Self-management Support System DialBetesPlus for Diabetic Kidney Disease: Protocol for a Randomized Controlled Trial. JMIR Res. Protoc. 10, e31061 (2021). https://www.researchprotocols.org/2021/8/e31061.
Menon, S. M. & Zink, R. C. Modern Approaches to Clinical Trials Using SAS : Classical, Adaptive, and Bayesian Methods. SAS Institute Inc., 2015.
Ajzen, I. & Schmidt, P. Changing Behavior using the Theory of Planned Behavior. In: Hagger, M., Cameron, L., Hamilton, K., Hankonen, N. & Lintunen T. eds. The Handbook of Behavior Change. Cambridge University Press. 2020:chap 2.
Leehey, D. J. et al. Aerobic exercise in obese diabetic patients with chronic kidney disease: a randomized and controlled pilot study. Cardiovasc. Diabetol. 8, 62 (2009).
Straznicky, N. E. et al. Exercise augments weight loss induced improvement in renal function in obese metabolic syndrome individuals. J. Hypertens. 29, 553–564 (2011).
Jakobsen, J. C. et al. When and how should multiple imputation be used for handling missing data in randomised clinical trials - a practical guide with flowcharts. BMC Med. Res. Methodol. 17, 162 (2017).
Graham, J. W. Missing data analysis: making it work in the real world. Annu. Rev. Psychol. 60, 549–576 (2009).
Hair, J. F. Jr., Black, W. C., Babin, B. J. eds. Multivariate Data Analysis. 7th ed. Pearson Education Ltd; 2010.
James, G., Witten, D., Hastie, T. eds. An introduction to statistical learning: Springer; 2013.
Trevor, H., Robert, T. & Ryan, T. Best Subset, Forward Stepwise or Lasso? Analysis and Recommendations Based on Extensive Comparisons. Statist. Sci. 35, 579–592 (2020).
Bassett, D. R. et al. Pedometermeasured physical activity and health behaviors in United States adults. Med. Sci. Sport Exerc. 42, 1819–1825 (2011).
Acknowledgements
We thank the patients and the participants from NIHON CHOUZAI Co., Ltd. and the eight participating institutions. This research was supported by AMED (grant JP19ek0210095). In addition to the AMED grant, this research was supported by NTT DOCOMO Inc. and Nihon Chouzai Co, Ltd. The funder played no role in the study design, data collection, analysis and interpretation of data, or the writing of this manuscript. ADDQoL use was licensed by the author, Clare Bradley (Health Psychology Research Unit, Royal Holloway, University of London).
Author information
Authors and Affiliations
Contributions
K.W. and M.N. contributed equally to this work. K.W. M.N. and E.K. drafted the manuscript, which was revised and approved by all authors, who also assume responsibility for its content. K.W., A.S., Y.K., K.M., H.W., Y.T., N.H., T.Y., S.K., A.I., S.S., T.M., T.I., T.T., Y.T., K.T., T.Y., M.N., N.K., and K.O. initiated the study. M.M. and S.E. performed the analyses, and all authors discussed the analysis plan and results and provided input to the manuscript. All authors had access to the final study results. K.W. is the guarantor of this work and, as such, has full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Corresponding author
Ethics declarations
Competing interests
K.W. has received research sponsorship/grants from NTT DOCOMO Inc. and Nihon Chouzai Co, Ltd.; and consulted for Nihon Chouzai Co, Ltd. During the study, K.W., A.S., and K.M. were members of a cooperative program between the University of Tokyo and NTT DOCOMO. The remaining authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Waki, K., Nara, M., Enomoto, S. et al. Effectiveness of DialBetesPlus, a self-management support system for diabetic kidney disease: Randomized controlled trial. npj Digit. Med. 7, 104 (2024). https://doi.org/10.1038/s41746-024-01114-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41746-024-01114-8
- Springer Nature Limited