DocG DocG | AI-Powered Clinical Summaries
Sign In

Stay Current with Medical Literature

AI-powered clinical summaries delivered to your inbox 3 times per week. Evidence-based insights in 300 words, designed for busy physicians.

Sign Up

See What You Get

Subscribers receive email summaries and access to our interactive AI-powered article rankings.

Email Summary Example

Online tai chi improves knee pain, function

A free, self-guided online tai chi program reduced walking knee pain and improved daily function more than online education alone after 12 weeks.
*Randomized clinical trial; Level 1b (OCEBM).

Citation

Zhu SJ, Hinman RS, Nelligan RK, et al. Online Unsupervised Tai Chi Intervention for Knee Pain and Function in People With Knee Osteoarthritis: The RETREAT Randomized Clinical Trial. JAMA Internal Medicine. 2026;186(1):15-25. Published online October 27, 2025. doi:10.1001/jamainternmed.2025.5723

Background

Tai chi is recommended for knee osteoarthritis, but in-person classes can be hard to access. This trial tested whether a fully self-guided online program can improve symptoms.

Patients

178 adults in Australia (average age ~62 years) with clinical knee osteoarthritis, knee pain for at least 3 months, and walking pain at least 4/10; required internet access. Key exclusions included regular exercise in the prior 3 months, recent knee surgery, inflammatory arthritis, or inability to walk unaided.

Intervention

Online program: education plus a 12-week video-based tai chi course (3 sessions/week) and encouragement to use a phone app that supports exercise habits.

Control

Education-only website about osteoarthritis and benefits of exercise.

Outcome

(Primary) Change in walking knee pain (0–10 scale) and difficulty with physical function (0–68 questionnaire).

Follow-up Period

12 weeks.

Results

Outcome (12 weeks) Difference vs control (95% confidence interval)
Walking knee pain (primary) −1.4 points (−2.1 to −0.7)
Physical function difficulty (primary) −5.6 points (−9.0 to −2.3)

Clinically noticeable improvement occurred more often with tai chi for pain (73% vs 47%; number needed to treat 4) and function (72% vs 52%; number needed to treat 6). No serious related harms were reported. Most participants met the program’s adherence target (82%).

Limitations

Participants reported their own outcomes and knew their group assignment, which can inflate perceived benefit. Follow-up was short (12 weeks). Average benefits were modest and slightly below the study’s prespecified “noticeable change” thresholds, although many more people reached those thresholds. The app and tai chi were tested together, so their separate effects are unclear. Results may not apply to people without reliable internet access.

Funding

Australian government medical research funding; no industry support reported.

Clinical Application

For knee osteoarthritis, consider prescribing this free online tai chi program when supervised classes are unavailable; it provides modest but meaningful 12-week symptom improvement.

Top Journal Rankings - April 2026

289 abstracts scored across 7 criteria. Click any article to expand criterion scores.
1. 8.1
Bleeding Risk with Apixaban vs. Rivaroxaban in Acute Venous Thromboembolism.
Overall: This large randomized trial provides fairly robust evidence that apixaban reduces clinically relevant bleeding versus rivaroxaban over 3 months in acute VTE, using outcomes that matter to patients and an intervention that is easy to implement, making it likely to influence real-world prescribing.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 7.5
Choice of DOAC for acute symptomatic PE/proximal DVT is a common outpatient-facing decision, though many initial VTE cases are started in ED/hospital or with specialist involvement rather than solely in primary care.
Validity, Bias Control & Precision 8.5
Large international randomized trial with blinded end-point assessment and clear effect estimate (RR 0.46 with tight 95% CI); open-label treatment could influence reporting/management of nonmajor bleeding, but endpoint definitions are standardized.
Patient-Oriented Outcomes 7.5
Primary outcome is clinically relevant bleeding (major or clinically relevant nonmajor), which is patient-important; mortality is reported but rare with very imprecise estimates.
Magnitude of Net Benefit 7.5
Clinically relevant bleeding was lower with apixaban (3.3% vs 7.1%; absolute reduction 3.8%), with no clear signal of offsetting serious non-bleeding harms in the abstract, though efficacy outcomes beyond mortality are not presented.
Implementability & Practicality 9.0
Both regimens use widely available standard-dose oral anticoagulants with familiar dosing and no special equipment, making adoption straightforward in routine outpatient care.
Practice-Changing Potential 8.5
A sizable reduction in clinically relevant bleeding in a large randomized trial is likely to shift clinician preference toward apixaban for many patients starting DOAC therapy for acute VTE.
2. 7.9
Maternal RSV vaccination for infant protection: A systematic review and meta-analysis of phase 3 trials with an integrated economic evaluation.
Overall: A large, well-conducted meta-analysis of phase 3 RCTs shows meaningful reductions in infant RSV and severe disease without major maternal/infant safety signals, with practical implementation largely dependent on vaccine cost and programmatic access.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 8.0
Maternal vaccination decisions commonly occur in outpatient prenatal care and can be relevant to primary care/family medicine workflows, though implementation is often coordinated with obstetrics and immunization programs.
Validity, Bias Control & Precision 9.0
Systematic review and meta-analysis restricted to phase 3 randomized placebo-controlled trials (n=17,391) with duplicate screening/extraction, RoB-2 assessment, random-effects pooling, and CIs reported, supporting credible and reasonably precise estimates.
Patient-Oriented Outcomes 8.0
Outcomes include infant RSV infection and severe RSV disease, which are clinically meaningful; however, key hard endpoints like hospitalization and mortality are not presented as pooled trial outcomes (mortality appears in modeling).
Magnitude of Net Benefit 7.5
Relative reductions are substantial (RR 0.47 for infection; RR 0.36 for severe disease) with no safety signals for major pregnancy outcomes, but absolute benefit is moderate (NNV 85 for infection; 127 for severe disease).
Implementability & Practicality 7.0
A single maternal vaccine dose is straightforward to deliver, but real-world uptake may be constrained by pricing, payer/public-sector purchasing, and program logistics highlighted by the economic scenarios.
Practice-Changing Potential 8.0
High-certainty phase 3 RCT evidence synthesized with safety data and policy-focused economic modeling could meaningfully support integration into national immunization strategies, though cost considerations may limit immediate universal adoption.
3. 7.3
Does a long time to colonoscopy after a positive faecal immunochemical test result have a deleterious impact on colorectal cancer outcomes? A nationwide cohort study.
Overall: A very large real-world cohort study highly relevant to FIT-based screening that suggests delays beyond 3 months may not worsen CRC stage outcomes up to 24 months, supporting practical triage by faecal haemoglobin level despite inherent limitations of retrospective observational evidence.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 9.0
Directly addresses a common primary-care screening pathway (FIT-positive follow-up) and informs real-world decisions about urgency and prioritization of colonoscopy after abnormal screening.
Validity, Bias Control & Precision 7.0
Very large nationwide retrospective cohort with adjusted analyses and reasonably tight confidence intervals, but nonrandomized design leaves substantial potential for confounding (e.g., sicker patients may be scoped sooner).
Patient-Oriented Outcomes 7.0
Reports clinically meaningful outcomes (CRC detection and advanced-stage CRC) rather than only test characteristics, though it does not report downstream outcomes like mortality or quality of life.
Magnitude of Net Benefit 6.0
Findings suggest no worse cancer/advanced-stage risk with longer intervals up to 24 months versus 2–3 months, implying potential system and patient burden reduction, but there is no demonstrated improvement in patient outcomes.
Implementability & Practicality 8.0
The implication (focus on ensuring completion and triage by faecal haemoglobin concentration rather than rigid timing targets) is feasible to apply in screening programs and primary-care workflows.
Practice-Changing Potential 7.0
Challenges strict time-to-colonoscopy targets and supports risk-based prioritization (higher f-Hb sooner), which could meaningfully influence program policy and referral urgency, though observational design may temper immediate change.
4. 7.3
Do Vitreous Floaters Predict Retinal Detachment? Retrospective Cohort Study in Primary Care.
Overall: A highly primary-care-relevant retrospective cohort linking floaters to retinal detachment risk with practical history-based stratification, but observational limitations and incomplete information on downstream harms/benefits constrain how decisively it should change referral practice.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 9.0
Directly addresses a common primary-care presentation (new-onset floaters/flashes) using data from seven family practices and informs the primary-care decision of urgency/referral for possible retinal detachment.
Validity, Bias Control & Precision 5.5
Retrospective cohort based on EHR review is vulnerable to misclassification and selection/documentation bias, with limited control of confounding described; precision is moderate with some wide confidence intervals despite a reasonable sample (n=1,181).
Patient-Oriented Outcomes 8.5
Retinal detachment is a clinically important, patient-centered outcome with major implications for vision and function; the study focuses on risk prediction for this hard endpoint rather than surrogate markers.
Magnitude of Net Benefit 6.5
Findings show clinically meaningful absolute risks (e.g., RD 6.1% with floaters alone vs 4.7% with flashes alone) and increased relative risks in higher-risk subgroups, but the abstract does not quantify downstream benefits/harms of acting on this information (e.g., referral burden, false positives).
Implementability & Practicality 8.0
Uses simple history features (new onset, acute ≤14 days, ‘many’ floaters/curtain) that are easy to assess in routine visits without special equipment, though it may increase urgent ophthalmology referrals.
Practice-Changing Potential 6.5
Could shift triage thresholds by elevating floaters as an alarm symptom (potentially above flashes), but observational design and limited detail on management pathways and outcomes of different referral strategies temper immediate practice change.
5. 7.3
Hormonal contraception and risk of major adverse cardiovascular events: A nationwide registry study from Finland.
Overall: A large national registry analysis addressing a common primary-care decision using hard cardiovascular outcomes, but observational design and wide confidence intervals temper certainty about small-to-moderate risk differences.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 9.0
Contraceptive prescribing and cardiovascular risk counseling are common, high-frequency primary-care decisions for women of reproductive age.
Validity, Bias Control & Precision 6.0
Very large nationwide registry with incident outcome capture, but the nested case-control comparison is observational (residual confounding likely) and the confidence intervals are wide with relatively few MACE events.
Patient-Oriented Outcomes 9.0
Outcomes are clearly patient-important hard endpoints (myocardial infarction and ischemic stroke), not surrogate markers.
Magnitude of Net Benefit 5.0
Point estimates suggest no meaningful increase in MACE risk versus nonuse, but uncertainty remains because CIs include potentially important harm; the study does not quantify other harms or benefits of contraception use.
Implementability & Practicality 9.0
Findings are straightforward to apply in routine contraceptive counseling without special tests, equipment, or added monitoring.
Practice-Changing Potential 6.0
Likely reassuring and may support current prescribing patterns, but observational design and imprecision limit its ability to change guidelines or decisively alter practice.
6. 7.1
Maternal supplementation of vitamin B
Overall: A reasonably strong blinded RCT shows a small improvement in infant mental developmental scores with higher-dose maternal B12 supplementation, with high practicality but limited harms reporting and modest effect size limiting the certainty and breadth of practice change.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 7.0
Antenatal/postpartum supplementation decisions are relevant to generalist maternity/primary care, but the study is in two tertiary centers and limited to first-trimester pregnant women following a vegetarian diet in India/Nepal, which narrows generalizability.
Validity, Bias Control & Precision 7.5
Double-blind randomized controlled trial with balanced baseline characteristics and a fairly large completed sample (531), but the abstract does not report allocation concealment, attrition from enrollment to completion, confidence intervals, or details on follow-up completeness.
Patient-Oriented Outcomes 8.0
Infant neurodevelopment (mental and motor developmental quotients at 9–12 months) is a patient-important outcome; however, longer-term functional outcomes are not reported and some outcomes mentioned (biochemical measures) are not fully described.
Magnitude of Net Benefit 5.0
The mental DQ difference is statistically significant but small in absolute terms (about a 2-point DQ difference / 7.8 centiles), motor DQ shows no significant difference, and the abstract provides no harms/adverse events or burden data to judge net benefit.
Implementability & Practicality 9.0
A daily oral methyl-cobalamin supplement is simple and low-complexity to prescribe and distribute in routine outpatient prenatal/postnatal care, with no specialized equipment or procedures described.
Practice-Changing Potential 6.0
Findings suggest a modest improvement in infant mental development with higher-dose supplementation in a defined risk group (vegetarian mothers), but limited effect size, unclear safety reporting, and context-specific setting make immediate broad practice change less certain.
7. 7.0
Nirsevimab vs RSVpreF Vaccine for Respiratory Syncytial Virus-Related Hospitalization in Newborns.
Overall: A large, well-analyzed real-world cohort shows nirsevimab is associated with fewer RSV hospitalizations and severe outcomes than maternal RSVpreF vaccination, with patient-important endpoints but only modest apparent absolute benefit and residual confounding risk.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 8.0
Addresses prevention of RSV hospitalization in newborns, a common concern in outpatient pediatrics/primary care and perinatal care, comparing two real-world immunization strategies.
Validity, Bias Control & Precision 7.0
Large population-based matched cohort with conditional Cox models and inverse probability weighting improves comparability, but the nonrandomized design leaves potential residual confounding; estimates are reasonably precise with reported CIs.
Patient-Oriented Outcomes 9.0
Primary and secondary outcomes are clearly patient-important (RSV-related hospitalization, PICU admission, ventilator support, oxygen therapy, death as censoring), not merely surrogate markers.
Magnitude of Net Benefit 5.0
Relative risk reductions are meaningful, but the absolute event rates appear low (hospitalizations ~1% in each group), suggesting a modest absolute benefit; harms, adverse events, and patient/family burden are not reported.
Implementability & Practicality 7.0
Both strategies are already in clinical use; nirsevimab before discharge is operationally feasible but may involve access, cost, and coordination barriers not described in the abstract.
Practice-Changing Potential 6.0
Findings suggest nirsevimab may outperform maternal RSVpreF vaccination for preventing RSV hospitalizations and severe outcomes, but observational evidence and modest absolute differences limit immediate definitive practice change.
8. 7.0
Acetaminophen (Paracetamol) or Opioid Analgesia Added to Ibuprofen for Children's Musculoskeletal Injury: Two Randomized Clinical Trials.
Overall: Well-designed pediatric RCT evidence shows no added short-term pain benefit from adding acetaminophen or hydromorphone to ibuprofen, with higher adverse events for hydromorphone, making the main impact a clear signal to avoid routine opioid add-on despite strong trial methods.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 7.0
Acute pediatric musculoskeletal injury pain control is a common outpatient/urgent-care decision, but the study population and workflow are from tertiary pediatric emergency departments, which may differ from typical primary-care settings.
Validity, Bias Control & Precision 8.5
Two randomized, double-masked, placebo-controlled multicenter trials with prespecified primary efficacy and safety outcomes and moderate sample sizes; however, the abstract provides limited precision metrics (no confidence intervals) and excludes some randomized participants from efficacy analyses.
Patient-Oriented Outcomes 8.0
Pain scores (patient-reported) and adverse events are patient-important outcomes, though pain is assessed only at 60 minutes and does not address longer-term function, return to activity, or subsequent care use.
Magnitude of Net Benefit 2.0
There was no improvement in pain with added acetaminophen or hydromorphone versus ibuprofen alone (non-significant difference), while adverse events were substantially higher with hydromorphone, yielding an unfavorable benefit–harm profile for add-on opioid therapy.
Implementability & Practicality 9.0
The compared regimens are simple single oral doses of commonly available medications with straightforward dosing, making the findings easy to apply where these drugs are used.
Practice-Changing Potential 7.5
The results directly inform a common analgesic escalation question and provide strong evidence against routine addition of acetaminophen or hydromorphone to ibuprofen for short-term pain relief, particularly discouraging opioid use due to higher adverse events.
9. 6.9
Electronic Health Record Intervention and Deprescribing for Older Adults: A Randomized Clinical Trial.
Overall: A well-conducted cluster RCT in primary care showing EHR nudges increase deprescribing, but the main endpoint is a process measure and patient-level clinical benefit remains unproven.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 9.0
Targets deprescribing of commonly problematic medications in adults ≥65 within primary care visits, directly aligning with routine generalist prescribing and medication review decisions.
Validity, Bias Control & Precision 8.5
Cluster randomized clinical trial with appropriate adjustment for clustering and multiple testing; large patient sample and reported CIs support reasonably precise estimates, though conducted in a single academic center and outcome ascertainment is EHR-based.
Patient-Oriented Outcomes 4.0
Primary outcome is deprescribing (a care process) rather than downstream patient outcomes (falls, cognition, function, quality of life); limited safety data are provided (no serious adverse events reported; death rates listed without clear interpretation).
Magnitude of Net Benefit 6.0
Shows a meaningful absolute increase in deprescribing (10.4% precommitment; 6.5% boostering vs usual care) with no serious adverse events reported, but net patient benefit is uncertain because clinical outcomes from deprescribing are not measured.
Implementability & Practicality 7.5
Intervention is low-intensity (EHR messages/reminders) and could be integrated into routine workflows, but requires EHR build capability and may not translate seamlessly across health systems with different EHR configurations.
Practice-Changing Potential 6.5
Provides actionable evidence that behavioral-science-designed EHR nudges can increase deprescribing in primary care, but lack of demonstrated improvement in patient-centered outcomes may limit immediate widespread practice change.
10. 6.9
Coffee and Tea Intake, Dementia Risk, and Cognitive Function.
Overall: A large, well-conducted observational analysis links moderate caffeinated coffee/tea intake to lower dementia risk, but residual confounding and limited harm reporting reduce confidence for practice-changing recommendations despite high primary-care relevance.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 8.0
Coffee/tea intake is a common primary-care lifestyle discussion, and dementia prevention/cognitive health is broadly relevant to generalist counseling.
Validity, Bias Control & Precision 6.5
Very large, long-running prospective cohorts with repeated dietary measures and multivariable adjustment, but the observational design leaves substantial potential for residual confounding and exposure misclassification; dementia ascertainment relies on records/diagnoses rather than uniform adjudication in the abstract.
Patient-Oriented Outcomes 8.5
Incident dementia is a patient-important outcome; subjective cognitive decline and telephone neuropsychological tests are clinically relevant but less definitive than hard outcomes, and objective testing was limited to one cohort.
Magnitude of Net Benefit 5.5
Associations suggest a moderate reduction in dementia risk (HR 0.82 with reported absolute incidence rates), but cognitive score differences are small and harms/burdens of increased caffeine intake are not reported in the abstract, limiting net-benefit assessment.
Implementability & Practicality 9.0
If acted on, recommending moderate coffee/tea intake is low-cost and easy to implement without specialized resources.
Practice-Changing Potential 4.0
Findings are association-based rather than causal, so they are more hypothesis-supporting than directive; unlikely to justify changing clinical recommendations on their own.
Score Guide: 9-10 Exceptional 7-8 Strong 5-6 Moderate 3-4 Weak 1-2 Poor
Showing top 10 of 289

How It Works

1

Sign Up with Google

Create your 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

Find the Best New Articles

Our interactive article rankings allow you to find the best articles, tailored to you.

Ready to Stay Current?

Sign Up