DocG DocG | AI-Powered Clinical Summaries
Sign In

Stay Current with Medical Literature

AI-powered clinical summaries delivered to your inbox 3x per week. Evidence-based insights in 250 words, designed for busy physicians.

Get Started Free

See What You Get

Free members receive email summaries. Premium members get web access plus AI-powered article rankings.

Email Summary Example

Free Tier
Free tier: Summaries delivered to your inbox only. Upgrade to view on web, search the archive, and customize email frequency.
Automated app coaching matched human diabetes prevention
Referring adults with prediabetes to a fully automated, app-based lifestyle program achieved similar 12-month health targets as referral to human coaching.
*Pragmatic noninferiority randomized clinical trial; Level 1b (OCEBM).

Citation

Mathioudakis N, Lalani B, Abusamaan MS, et al; AI-DPP Study Group. An artificial intelligence–powered lifestyle intervention vs human coaching in the Diabetes Prevention Program: a randomized clinical trial. JAMA. 2025;334(23):2079-2089. doi:10.1001/jama.2025.19563

Background

Prediabetes is common, but structured lifestyle programs are underused because of limited capacity and participation barriers. This trial tested whether an automated, app-based program could perform at least as well as standard remote human coaching.

Patients

368 adults (median age 58 years) with prediabetes and overweight or obesity at 2 US sites.

Intervention

Referral to a fully automated, app-based Diabetes Prevention Program with a Bluetooth scale.

Control

Referral to a remote, group-based, human coach–led Diabetes Prevention Program.

Outcome

(Primary) Composite at 12 months: maintained hemoglobin A1c below 6.5% throughout study and met ≥1 of: ≥5% weight loss; or ≥4% weight loss plus ≥150 weekly minutes of physical activity (measured by wearable); or hemoglobin A1c drop ≥0.2%.

Follow-up Period

12 months

Results

Primary analysis was intention-to-treat; participants missing 12-month data were counted as not meeting the outcome.
Outcome Automated app referral Human-coach referral Difference
Primary composite outcome (primary) 31.7% (58/183) 31.9% (59/185) −0.2%; noninferior (margin −15%)
Program initiation 93.4% 82.7% +10.7%; NNT 10
Program completion 63.9% 50.3% +13.6%; NNT 8

Limitations

Unblinded; used short-term targets rather than diabetes onset. Human coaching was remote (not in-person). Two sites with motivated participants may limit generalizability.

Funding

National Institutes of Health grants; intervention vendors paid; no funder role.

Clinical Application

When access to human coaching is limited, clinicians can refer eligible patients to a fully automated app-based program and expect similar 12-month results.

Top Journal Rankings - February 2026

Premium
28 abstracts scored across 7 criteria. Click any article to expand criterion scores.
1. 8.7
SGLT2 Inhibitors and Kidney Outcomes by Glomerular Filtration Rate and Albuminuria: A Meta-Analysis.
Overall: A large, rigorous meta-analysis of blinded RCTs shows clinically important reductions in CKD progression and kidney failure with SGLT2 inhibitors across the full range of baseline eGFR and albuminuria, supporting broader routine use with typical outpatient monitoring.
View 6 Criterion Scores
Relevant 9.0
Addresses CKD progression risk reduction with SGLT2 inhibitors across eGFR/albuminuria strata in common primary-care populations (type 2 diabetes, CKD, heart failure), including stage 4 CKD and minimal albuminuria.
Valid 9.5
Large inverse-variance meta-analysis of 10 randomized, double-blind, placebo-controlled trials (70,361 participants) with prespecified kidney outcomes and precise effect estimates (HRs with 95% CIs).
Change in Practice 8.0
Directly informs a common clinical uncertainty (use in eGFR <30 and low UACR) and supports broader routine use, potentially reducing under-prescribing in advanced CKD/low albuminuria.
Applicable to Medical Care 8.5
Intervention is standard pharmacotherapy and outcomes are clinically relevant; real-world use requires routine labs and attention to CKD comorbidity management but no specialized infrastructure is implied.
Implementable 8.0
SGLT2 inhibitors are widely available and can be prescribed in typical outpatient care, though safe adoption often requires monitoring (renal function/volume status) and coordination in advanced CKD.
Clinically Meaningful 9.0
Demonstrates benefit on patient-important kidney outcomes (CKD progression and kidney failure) with substantial relative risk reduction (CKD progression HR 0.62; kidney failure HR 0.66) and reported absolute event rates per 1000 patient-years.
2. 8.2
Caffeinated Coffee Consumption or Abstinence to Reduce Atrial Fibrillation: The DECAF Randomized Clinical Trial.
Overall: A multicenter randomized trial found lower AF/atrial flutter recurrence with modest caffeinated coffee intake after cardioversion, using a practical and easily implementable intervention, though open-label design and limited reporting of bias safeguards temper certainty.
View 6 Criterion Scores
Relevant 8.0
Atrial fibrillation management and lifestyle counseling are common in generalist care, though the study population is a post-cardioversion, largely male, multicenter cohort rather than an explicitly primary-care setting.
Valid 7.5
Prospective multicenter randomized trial with a prespecified primary endpoint and trial registration, but it was open-label and the abstract does not report allocation concealment, blinding of outcome assessment, or intention-to-treat handling.
Change in Practice 7.0
Directly tests a common counseling recommendation (caffeine/coffee avoidance) and suggests benefit, but the finding is limited to coffee drinkers after successful cardioversion and may need replication before broadly shifting guidance.
Applicable to Medical Care 9.0
The intervention is simple dietary guidance (encourage ~1 cup/day vs abstinence) without specialized equipment or monitoring beyond usual AF follow-up.
Implementable 9.5
Coffee consumption counseling can be adopted immediately in routine visits using readily available products and minimal workflow changes.
Clinically Meaningful 8.5
Shows a patient-important outcome (AF/atrial flutter recurrence) with a substantial absolute difference (47% vs 64%) and a significant hazard ratio (0.61; 95% CI, 0.42-0.89), with no significant adverse event difference reported.
3. 8.2
Effects of Sodium Glucose Cotransporter 2 Inhibitors by Diabetes Status and Level of Albuminuria: A Meta-Analysis.
Overall: This large meta-analysis of randomized trials reports consistent, clinically important reductions in kidney progression, AKI, and hospitalization (and some mortality benefit), supporting broad CKD use of SGLT2 inhibitors with generally feasible implementation in routine care.
View 6 Criterion Scores
Relevant 8.5
Chronic kidney disease and diabetes/non-diabetes CKD management are common in general medical care, and outcomes (kidney progression, AKI, hospitalization, death) are directly relevant to typical primary-care decision-making.
Valid 9.0
Inverse-variance meta-analysis of 8 randomized clinical trials with a very large sample and consistent effect estimates with 95% CIs; however, the abstract does not report risk-of-bias assessment or protocol registration details.
Change in Practice 7.5
Findings may strengthen or broaden confidence in using SGLT2 inhibitors for CKD irrespective of diabetes status and across albuminuria strata, but it is partly confirmatory of an already-established drug class rather than a wholly new approach.
Applicable to Medical Care 8.0
Treatment is medication-based and generally manageable in routine care with standard lab monitoring; no specialized procedures are required, though CKD prescribing/monitoring adds some workflow complexity.
Implementable 7.5
SGLT2 inhibitors are already available and can be started in clinical practice, but real-world uptake can be limited by contraindication screening, follow-up needs, and coverage/cost issues not addressed in the abstract.
Clinically Meaningful 8.5
Shows absolute event-rate reductions for kidney disease progression, AKI, and hospitalization, with mortality benefit clearer in diabetes than non-diabetes (CI crosses 1.0 in non-diabetes), supporting an overall favorable patient-important benefit profile.
4. 7.5
Effects of lifestyle interventions in pregnancy on gestational diabetes: individual participant data and network meta-analysis.
Overall: A large, registered IPD/network meta-analysis of randomized trials shows modest, criterion-dependent reductions in gestational diabetes with lifestyle interventions, highly credible but only moderately likely to change practice given implementation demands and variable effects by diagnostic definition.
View 6 Criterion Scores
Relevant 9.0
Focuses on preventing gestational diabetes during pregnancy—a common condition managed in obstetric and family medicine settings—with broad inclusion across many randomized trials.
Valid 9.0
Large individual participant data and network meta-analysis of randomized trials (104 trials; ~36k women), includes confidence intervals, prespecified outcomes, and PROSPERO registration; still some uncertainty due to varying diagnostic criteria and mixed IPD availability.
Change in Practice 6.0
Supports and refines existing lifestyle counseling (with criterion-dependent effects) but the modest absolute risk reductions and lack of effect under NICE criteria make it less clearly practice-changing on its own.
Applicable to Medical Care 7.0
Diet/physical-activity interventions are clinically deliverable but often require structured programs, time, and potential training/group formats, which can be challenging in routine prenatal care workflows.
Implementable 7.0
Can be adopted using standard counseling and activity guidance immediately, though achieving effective, equitable delivery may need additional implementation supports (e.g., provider training, group programs).
Clinically Meaningful 7.0
Shows modest but potentially important reductions in gestational diabetes with absolute risk reductions ~2.6–3.5% (depending on criteria) and CIs provided; benefit is less clear when using NICE criteria (no observed reduction).
5. 6.8
Provision of knee bracing for knee osteoarthritis (PROP OA): multicentre, parallel group, superiority, statistician blinded, randomised controlled trial.
Overall: A well-conducted multicentre RCT in a primary-care-relevant condition found small, time-limited improvements in patient-reported outcomes with knee bracing plus adherence support, with feasible but non-trivial implementation requirements.
View 6 Criterion Scores
Relevant 8.5
Knee osteoarthritis is very common in primary care, and recruitment occurred through general practices/community, making the population and problem closely aligned with generalist care.
Valid 8.8
Multicentre randomized controlled trial with centralized web-based randomization and statistician blinding; sample size is substantial (n=466) with reasonable follow-up, though attrition increases by 12 months.
Change in Practice 5.5
Shows statistically significant but small improvements versus advice/information/exercise alone, with benefits diminishing over time, making major practice change less likely on the basis of this effect size.
Applicable to Medical Care 6.5
Intervention requires compartment-specific brace selection/fitting by a trained physiotherapist plus adherence supports and a follow-up visit, which is feasible but not universally simple in routine workflows.
Implementable 6.0
Braces and motivational/text adherence supports are available now, but implementation needs access to appropriate braces, fitting expertise, and a structured follow-up process.
Clinically Meaningful 5.8
Patient-reported outcomes improved with modest absolute differences (KOOS-5 mean difference 3.39; pain subscale 6.13) and minor expected adverse events; the net benefit appears small based on reported effect sizes.
6. 6.7
An mHealth (Mobile Health) Intervention for Smoking Cessation in People With Tuberculosis: A Cluster Randomized Clinical Trial.
Overall: A well-conducted multicenter cluster RCT shows large, biochemically verified improvements in smoking cessation with an SMS-based program in TB patients, with practical (though not turnkey) implementation considerations and more limited direct generalizability to typical U.S. primary care.
View 6 Criterion Scores
Relevant 5.0
Addresses smoking cessation (common primary-care concern) but the population is people with drug-sensitive pulmonary TB in TB clinics in Bangladesh/Pakistan, which is less directly representative of typical U.S. primary-care settings.
Valid 8.0
Multicenter cluster randomized clinical trial with prespecified primary outcome, good retention (91%), trial registration, and biochemical verification of abstinence; some design details (e.g., blinding/cluster-level analytic handling) are not described in the abstract.
Change in Practice 6.0
Shows a large improvement in verified abstinence with a low-intensity intervention, which could reasonably prompt adoption of SMS support, though it is not a head-to-head against intensive pharmacotherapy/counseling and is in a TB-specific context.
Applicable to Medical Care 7.0
Text-message support is generally compatible with routine outpatient workflows and does not require specialized equipment beyond phone access, but it does require an organized messaging program and patient phone availability.
Implementable 6.0
In principle could be deployed with existing SMS tools, but the intervention requires setting up message schedules/content and integrating it into care (not a simple one-click change described for immediate use).
Clinically Meaningful 8.0
Primary outcome is patient-important and objectively verified: 41.7% vs 15.3% continuous abstinence at 6 months (RR 3.0; 95% CI 2.0-4.9; absolute difference 26.4%); mortality was lower (HR 0.4; 95% CI 0.2-0.9) but appears secondary and adverse effects are not reported.
7. 6.2
DASH-Patterned Groceries and Effects on Blood Pressure: The GoFresh Randomized Clinical Trial.
Overall: A well-conducted community RCT shows modest short-term BP and LDL improvements from delivered DASH groceries plus counseling, but limited durability and substantial delivery/support requirements reduce near-term clinical uptake.
View 6 Criterion Scores
Relevant 8.0
Addresses elevated blood pressure in untreated adults in a community setting—highly pertinent to primary care prevention and early hypertension management, though restricted to Black residents of specific urban communities.
Valid 8.5
Parallel-group randomized clinical trial with a prespecified primary outcome, high completion (97%), objective adherence measure (24-hour urine), and effect estimates with CIs; blinding and other safeguards are not described.
Change in Practice 5.0
Shows modest additional BP lowering versus financial stipends, but effects were not maintained after stopping the program, limiting immediate impact on routine clinical decision-making.
Applicable to Medical Care 5.5
Dietitian counseling and dietary approaches are clinically relevant, but weekly ordering plus home-delivered groceries is not a typical clinic workflow and may depend on external program infrastructure.
Implementable 4.5
Not a simple point-of-care intervention; requires grocery delivery logistics and dietitian support, and the comparator included substantial stipends, making immediate adoption by most practices difficult.
Clinically Meaningful 6.0
Demonstrates a statistically significant additional systolic BP reduction (−3.4 mm Hg) and LDL reduction, but benefits are modest, durability is lacking, and harms/burdens are not detailed.
8. 6.0
Statin Use Is Not Associated With Reduced Cardio- and Cerebrovascular Hospitalizations in Older Adults With Dementia.
Overall: A very large observational cohort suggests a small increased hospitalization risk with statins in dementia, but limited causal certainty and modest effect size reduce confidence for practice change despite high real-world applicability.
View 6 Criterion Scores
Relevant 6.0
Addresses a common primary-care decision (statin use) in older adults, but the population is nursing home residents with/without dementia in Germany, which may differ from many U.S. primary-care patients.
Valid 4.5
Large retrospective claims-based cohort with propensity score methods, but observational design and limited abstract detail on confounding control, exposure definition, and outcome adjudication constrain causal inference.
Change in Practice 5.0
Findings may prompt more cautious prescribing/deprescribing in dementia, but the effect is small and nonrandomized evidence is unlikely to change standards on its own.
Applicable to Medical Care 8.0
Statin continuation or discontinuation is a routine clinical decision and does not require specialized equipment or infrastructure.
Implementable 8.0
Clinicians can act on the implication (reassess statin need in dementia) immediately using existing medications and workflows.
Clinically Meaningful 4.5
Outcome (cardio-/cerebrovascular hospitalization) is patient-important, but the reported association is small (HR 1.06, CI narrowly above 1.0) and may reflect residual confounding rather than a clear net harm/benefit.
9. 6.0
Efficacy and Safety of Oral PCSK9 Inhibitor Enlicitide in Adults With Heterozygous Familial Hypercholesterolemia: A Randomized Clinical Trial.
Overall: This well-conducted phase 3 RCT shows substantial LDL-C lowering and similar short-term safety with an oral PCSK9 inhibitor in HeFH, but the evidence is mainly surrogate-based, placebo-controlled, and not yet clearly ready for immediate routine use.
View 6 Criterion Scores
Relevant 6.0
Familial hypercholesterolemia and LDL-C management are commonly encountered in general practice, but the population is a specific high-risk subgroup (HeFH) often co-managed with lipid specialists; the abstract does not describe a primary-care setting.
Valid 8.0
Phase 3 randomized clinical trial with multicenter international enrollment, prespecified primary outcome at 24 weeks, high completion (96.7%), and clear effect estimates with 95% CIs; blinding and allocation concealment are not reported.
Change in Practice 6.0
Shows very large LDL-C lowering on top of background statins/ezetimibe, which could influence lipid-lowering strategies, but it is placebo-controlled (not compared with existing PCSK9 therapies) and focuses on surrogate lipid outcomes rather than events.
Applicable to Medical Care 8.0
A once-daily oral medication layered onto standard lipid therapy is operationally feasible in routine outpatient care with typical lab monitoring.
Implementable 3.0
The abstract presents a clinical trial of an oral PCSK9 inhibitor without stating it is approved or available for prescribing, limiting immediate adoption in typical practice.
Clinically Meaningful 5.0
Demonstrates large improvements in LDL-C and related lipids with similar adverse-event rates, but the outcomes are surrogate biomarkers and no cardiovascular events, quality of life, or absolute risk reductions are reported.
10. 6.0
Stress Hyperglycemia Ratio and Adverse Outcomes in Acute Mild Ischemic Stroke or High-Risk Transient Ischemic Attack: A Secondary Analysis of the INSPIRES Trial.
Overall: A large secondary analysis shows elevated stress hyperglycemia ratio predicts worse 90-day stroke-related outcomes, but it is primarily prognostic (not interventional), limiting immediate practice change despite being easy to compute from routine labs.
View 6 Criterion Scores
Relevant 6.5
Acute mild ischemic stroke/high-risk TIA is common and often first encountered in ED/inpatient general medicine, but the analysis targets atherosclerotic-cause patients and is less directly tied to routine outpatient primary-care decisions.
Valid 6.0
Large sample drawn from a randomized trial cohort with prespecified clinical outcomes and multivariable adjustment, but this is a secondary observational association analysis (SHR not randomized) with residual confounding risk and limited methodological safeguards described in the abstract.
Change in Practice 3.5
Findings mainly add prognostic risk stratification (higher SHR associated with worse outcomes) without testing an intervention or providing guidance that would clearly change standard management.
Applicable to Medical Care 7.5
Uses routinely available labs (admission glucose, HbA1c) and standard time horizons/outcomes, making it practical to apply in typical hospital workflows.
Implementable 7.5
SHR can be calculated immediately from common lab values without new equipment or therapies, though it would require adding a calculator/decision support step.
Clinically Meaningful 5.0
Associations are with patient-important outcomes (recurrent stroke, functional outcome) and include hazard ratios with CIs, but absolute risk differences by SHR level and any actionable benefit-harm impact are not provided.
Score Guide: 9-10 Exceptional 7-8 Strong 5-6 Moderate 3-4 Weak 1-2 Poor
Showing top 10 of 28

How It Works

1

Sign Up with Google

Create your free account in seconds using your Google account. No passwords to remember.

2

Receive Email Summaries

Get AI-curated clinical summaries delivered to your inbox three times per week (Monday, Wednesday, Friday).

3

Upgrade for Web Access

Want to search the archive, access summaries on the web, or customize your topics? Upgrade to Premium.

Ready to Stay Current?

Get Started Free