Evaluating the Effectiveness of a Smart Screening and Tiered Management Tool for Diabetic Kidney Disease in Primary Care: A Cluster Randomized Trial (SMART-DKD)

This cluster-randomized controlled trial assesses the effectiveness of an integrated intelligent digital plug-in for standardized screening and stratified clinical management of diabetic kidney disease (DKD) within primary care electronic medical record (HIS) systems, compared against routine guideline-based diabetes care without digital decision support. The study addresses widespread under-screening, inconsistent risk stratification and substandard long-term renal protection for adults living with type 2 diabetes at community and township healthcare facilities across China. Thirty primary care institutions (15 urban community health centers, 15 rural township health centers) will undergo 1:1 cluster randomization to intervention or control arms, with a total target enrollment of 3,000 participants with type 2 diabetes (1,500 per arm). All study procedures comply with the Declaration of Helsinki, Chinese GCP and national regulations for investigator-initiated clinical research.

Eligibility Criteria Institutional Inclusion Criteria Each participating primary care site must register ≥200 managed type 2 diabetes patients under national basic public health services; have internet-connected HIS terminals to operate the digital management tool; complete on-site laboratory capacity to perform UACR, serum creatinine, fasting glucose, HbA1c, lipid panel and urinalysis with validated internal and external quality control; possess standardized biospecimen collection, temporary storage and cold-chain transport protocols to deliver blood and urine samples to the central testing laboratory at the First Affiliated Hospital of Chongqing Medical University; and secure full written commitment from site leaders and clinicians to complete recruitment, intervention and follow-up per protocol. Institutions with a record of premature project withdrawal due to poor investigator compliance within the past two years are excluded.

Participant Inclusion Criteria Age 18 to 75 years inclusive, any gender; Confirmed type 2 diabetes mellitus per the 2024 Chinese Type 2 Diabetes Prevention and Treatment Guidelines; Able to provide written informed consent voluntarily. Participant Exclusion Criteria Active severe systemic illness or acute disease flare at screening; Moderate-to-severe psychiatric disorders including depression, bipolar disorder, schizophrenia, or documented suicidal ideation; Cognitive impairment or language barriers preventing full understanding of study procedures and communication with research staff; Concurrent enrollment in another DKD screening or interventional research trial; Any other condition judged by investigators to preclude safe trial participation.

Participant Early Withdrawal Rules Participants may withdraw consent and exit the trial at any time without penalty to routine medical care. Mandatory withdrawal triggers include participant death, identification of non-diabetic kidney disease during follow-up, persistent severe non-adherence disrupting endpoint assessment, or three consecutive failed follow-up contacts within one month. All withdrawn participants will complete a standardized exit assessment within six weeks of discontinuation (exempt if full lab testing was completed within the prior four weeks), with all withdrawal reasons fully documented in case report forms.

Intervention and Control Definitions The intervention arm deploys a customized DKD screening and graded management plug-in embedded in local HIS. The digital tool auto-calculates eGFR, generates standardized DKD risk stratification, delivers automated UACR screening reminders, and provides individualized guidance on glycemic/blood pressure targets, medication adjustment, follow-up scheduling and specialist referral thresholds. Control sites deliver conventional diabetes management following national primary diabetes guidelines and public health chronic disease protocols, with no access to the intelligent digital module, automated renal function calculations or clinical alert prompts. All participating clinicians receive unified baseline training on general diabetes care; only intervention-site staff receive additional training on the digital tool's functions.

Trial Visit Schedule and Assessments Screening/baseline visit (V1, -2 to 0 weeks) confirms eligibility, collects informed consent, captures demographics, medical/surgical history, chronic medication records, standardized physical measurements, and baseline blood/urine biospecimens for central laboratory testing of UACR, serum creatinine, urinalysis, fasting glucose, HbA1c and lipids. Mandatory follow-up visits occur at 3 months (V2 ±15 days), 6 months (V3 ±15 days) and 12 months (V4 ±15 days), repeating identical physical exams, laboratory testing, adverse event documentation and medication reconciliation. An extended observational phase runs through Month 24 with repeated biomarker and clinical data collection. Unscheduled visits are permitted for participants with abnormal lab values or clinically significant adverse events, with all encounters fully recorded. All baseline, 12-month and 24-month biospecimens are shipped to the central research laboratory for unified testing.

Primary and Secondary Study Endpoints Primary Outcome Between-group difference in natural log-transformed urine albumin-creatinine ratio (UACR) from baseline to the 12-month follow-up timepoint, analyzed via repeated measures mixed-effects models (MMRM) across all longitudinal visits. Missing primary endpoint data will be handled via multiple imputation.

Secondary Outcomes (all assessed at 12 months post-baseline) Institutional DKD screening rate among enrolled type 2 diabetes participants; Confirmed DKD clinical diagnosis rate; Rate of guideline-concordant standardized DKD treatment; Change in estimated glomerular filtration rate (eGFR) from baseline to 12 months; Real-world operability of the intelligent digital tool in primary care and participant long-term follow-up adherence.

Safety Endpoints All adverse events will be coded per MedDRA version 28.0 or newer, including hyperkalemia, acute kidney injury, severe hypoglycemia and other clinically significant laboratory abnormalities. Incidence rates of adverse events are compared between groups using Fisher's exact test to identify safety signals; all serious adverse events undergo detailed causality assessment, narrative documentation and standardized reporting.

Analysis Populations

Four distinct analytical datasets will be generated for outcome evaluation:

Intention-to-Treat (ITT): All cluster-randomized participants, primary population for main efficacy analysis; Modified ITT (mITT): ITT participants who completed baseline assessment and at least one DKD screening visit, used for sensitivity analyses of efficacy endpoints; Per-Protocol (PPS): mITT participants with complete 12-month UACR/eGFR data and no major protocol violations, for robustness testing of primary outcomes; Safety Set (SS): All participants with at least one safety evaluation visit, used for all adverse event analyses.

Continuous secondary endpoints will be analyzed with t-tests or Wilcoxon rank-sum tests based on normality; categorical screening, diagnosis and treatment rates use Fisher's exact test. All statistical testing employs a two-sided alpha of 0.05, with analyses performed using R or SAS software.

Risks, Benefits and Data Confidentiality Minimal procedural risks to participants include transient pain, bruising or rare vasovagal reactions from routine venipuncture and urine collection. All trial-related laboratory testing is provided free of charge by the research team with no participant out-of-pocket costs. Intervention arm participants receive individualized digital renal risk management recommendations to slow DKD progression. Any study-related injury qualifies participants for free clinical treatment and project-funded compensation; no monetary incentives are offered for enrollment or follow-up. All participant identifiers are anonymized via unique study codes, with encrypted electronic databases and locked physical archives limiting data access to designated research and audit staff only. De-identified aggregate study results will be published in peer-reviewed journals and presented at academic conferences without disclosure of individual participant information.