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

Conversational artificial intelligence improved anxiety and well-being
Among distressed university students, a conversational artificial intelligence program reduced anxiety and improved well-being more than group therapy or a wait list.
*Randomized clinical trial; Level 1b (OCEBM).

Citation

Shoshani A, Gurfinkel B, Kor A, et al. Efficacy of a conversational artificial intelligence agent for psychiatric symptoms and digital therapeutic alliance: a randomized clinical trial. JAMA Network Open. 2026;9(4):e266713. doi:10.1001/jamanetworkopen.2026.6713

Background

Scalable mental health support is needed where clinician time is limited. Conversational artificial intelligence tools may increase access, but comparative evidence and how “working relationship” affects outcomes are limited.

Patients

995 university students (Israel), ages 18-35, reporting current distress. Excluded: active suicidal thoughts or crisis, current psychotherapy or psychiatric medication, and history of severe mental disorders.

Intervention

12-week conversational artificial intelligence mental health platform with on-demand messaging and coping exercises.

Control

Face-to-face group therapy (12 weekly 90-minute sessions) or wait list.

Outcome

Anxiety (primary), depression (primary), posttraumatic stress symptoms (primary), well-being (primary), life satisfaction (primary); plus engagement and perceived working relationship with the tool.

Follow-up Period

12 weeks (primary endpoint) plus 3 months.

Results

Outcome Comparison Adjusted mean difference (95% confidence interval)
Anxiety score (primary) Conversational artificial intelligence vs group therapy -2.17 (-2.67 to -1.67)
Anxiety score (primary) Conversational artificial intelligence vs wait list -2.15 (-2.65 to -1.65)
Depression score (primary) Conversational artificial intelligence vs wait list -1.99 (-2.63 to -1.35)
Well-being score (primary) Conversational artificial intelligence vs group therapy +5.72 (+2.71 to +8.73)
Well-being score (primary) Conversational artificial intelligence vs wait list +9.16 (+6.14 to +12.18)
Life satisfaction (primary) Conversational artificial intelligence vs wait list +2.58 (+1.62 to +3.54)
Posttraumatic stress symptoms: no meaningful differences between groups.
Clinical improvement (moved from above to below common symptom cutoffs at 12 weeks): anxiety—57.9% with conversational artificial intelligence vs 9.8% wait list (number needed to treat 2); depression—47.0% vs 14.0% (number needed to treat 4).
Analyses were intention-to-treat. No serious adverse events were reported.

Limitations

Students with severe illness, suicidality, and current treatment were excluded, limiting generalizability. Outcomes were self-reported and participants were unblinded. Benefits were modest and not seen for posttraumatic stress symptoms. Follow-up loss was about 35%. Some authors had financial ties to the platform company.

Funding

Not reported; some authors employed by or held equity in the platform company.

Clinical Application

Consider conversational artificial intelligence as an adjunct or early support for mild-to-moderate distress; do not rely on it for trauma-related symptoms or higher-risk patients.

Top Journal Rankings - June 2026

278 abstracts scored across 7 criteria. Click any article to expand criterion scores.
1. 8.1
Ensitrelvir for Covid-19 Postexposure Prophylaxis in Household Contacts.
Overall: A large, well-conducted blinded RCT shows ensitrelvir started quickly after household exposure substantially reduces symptomatic, PCR-confirmed Covid-19 with no evident safety penalty, making it promising for outpatient prevention if access and logistics allow.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 8.5
Addresses a common outpatient/household scenario (postexposure prophylaxis for Covid-19) with an intervention and timing window that primary care and urgent care commonly manage.
Validity, Bias Control & Precision 9.0
Large double-blind randomized placebo-controlled trial with a clearly defined primary endpoint, modified intention-to-treat analysis, and a precise effect estimate (RR 0.33; 95% CI 0.22–0.49).
Patient-Oriented Outcomes 8.0
Primary outcome requires both lab confirmation and symptomatic illness lasting ≥48 hours, which is patient-relevant; however, severe outcomes (hospitalization/death) did not occur, limiting assessment of harder endpoints.
Magnitude of Net Benefit 8.5
Meaningful absolute reduction in symptomatic Covid-19 (2.9% vs 9.0%; absolute risk reduction 6.1%, NNT ≈ 16) with similar adverse-event and serious adverse-event rates between groups.
Implementability & Practicality 7.0
Oral 5-day regimen is practical, but implementation requires rapid identification/testing of household contacts and initiation within 72 hours, which may be logistically challenging; drug availability/approval is not established in the abstract.
Practice-Changing Potential 7.5
Provides strong evidence for an antiviral postexposure prophylaxis strategy in households—a clear new use case—though broader practice change depends on access, regulatory status, and positioning versus existing prevention approaches.
2. 7.9
Efficacy and Safety of an mRNA Seasonal Influenza Vaccine in Adults.
Overall: This large, double-blind phase 3 trial in adults ≥50 found mRNA-1010 reduced RT-PCR–confirmed influenza-like illness versus a standard-dose licensed vaccine (2.0% vs 2.8%; relative vaccine efficacy 26.6%, 95% CI 16.7–35.4), providing precise and credible evidence for a patient-relevant preventive outcome. Benefits are modest in absolute terms and come with clearly higher short-term reactogenicity (e.g., injection-site pain, fatigue), while serious adverse events were similar. If the product beco
View 6 Criterion Scores
Primary-Care Relevance & Applicability 8.5
Targets a very common primary-care preventive service (influenza vaccination) in adults ≥50, with outcomes relevant to routine outpatient care.
Validity, Bias Control & Precision 9.0
Large phase 3 randomized, double-blind, active-controlled trial (n≈40,700) with clear primary endpoint and reported 95% CI, supporting strong internal validity and precision.
Patient-Oriented Outcomes 7.5
Primary outcome is RT-PCR–confirmed, protocol-defined influenza-like illness, which is clinically meaningful, though it is not a hard outcome like hospitalization or mortality.
Magnitude of Net Benefit 6.5
Absolute risk reduction for the primary outcome is modest (2.8% to 2.0%; ~0.8%); reactogenicity is substantially higher, while serious adverse events are similar.
Implementability & Practicality 7.5
Vaccination is highly implementable in primary care, but this is an investigational product and the abstract does not address availability, cost, or workflow considerations beyond administration.
Practice-Changing Potential 8.5
Shows statistically robust superiority versus a licensed standard-dose comparator in a large adult population, which could plausibly influence vaccine selection if adopted and made available.
3. 7.6
Colorectal Cancer and Mortality Risk Among Older Adults With vs Without Adenoma on Prior Colonoscopy.
Overall: A very large cohort provides precise, patient-important 10-year risk estimates showing only small absolute increases in CRC/CRC death after prior adenoma in older adults, with competing non-CRC mortality far higher, making it useful for primary-care shared decision-making despite observational limitations and VA male-dominant generalizability concerns.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 8.5
Directly informs a common primary-care decision in older adults: whether to continue/deprioritize surveillance colonoscopy after age 75, with competing-risk context.
Validity, Bias Control & Precision 7.5
Large retrospective VA cohort (n=91,952) with 10-year cumulative incidence estimates and CIs; however, observational design leaves room for confounding/selection bias and the population is predominantly male VA patients.
Patient-Oriented Outcomes 9.0
Reports patient-important outcomes including colorectal cancer incidence, colorectal cancer death, non-cancer death, and all-cause mortality over 10 years.
Magnitude of Net Benefit 5.5
Absolute differences are small (e.g., CRC 1.1% vs 0.7%; CRC death 0.5% vs 0.4% at 10 years) and are dwarfed by very high competing non-CRC mortality, suggesting limited potential net gain from aggressive surveillance in many.
Implementability & Practicality 7.5
Findings are readily usable in shared decision-making and prioritization discussions; frailty stratification could be applied if the VA Frailty Index (or similar) is available, though it may not be universally integrated in all primary-care workflows.
Practice-Changing Potential 7.5
Supports de-emphasizing surveillance colonoscopy in many adults ≥75 even with prior adenoma by quantifying low CRC risk versus high competing mortality, which could shift counseling and referral patterns, though it is not randomized evidence.
4. 7.6
Amoxicillin-Clavulanate vs Amoxicillin for Acute Sinusitis in Adults.
Overall: This very large outpatient database study directly informs a common primary-care prescribing choice and finds no difference in treatment failure between amoxicillin-clavulanate and amoxicillin, with increased (generally rare) secondary infections including yeast and C difficile for amoxicillin-clavulanate. Precision is good, but the retrospective observational design leaves potential residual confounding. Outcomes are clinically relevant though partly utilization-based. The practical takeaway—pr
View 6 Criterion Scores
Primary-Care Relevance & Applicability 9.5
Addresses a very common outpatient primary-care decision (choice of first-line antibiotic for uncomplicated acute sinusitis) in adults 18–64 using real-world outpatient data.
Validity, Bias Control & Precision 7.0
Large nationwide new-user, active-comparator cohort with propensity-score matching and sensitivity analyses improves credibility, but retrospective observational design leaves residual confounding possible; estimates are fairly precise (narrow CI around RR ~1.0).
Patient-Oriented Outcomes 7.0
Outcomes include treatment failure requiring new antibiotics/visits and ED/inpatient encounters plus adverse events and secondary infections (including C difficile), which are clinically meaningful, though the composite definition includes utilization-based components.
Magnitude of Net Benefit 6.5
No meaningful benefit for amoxicillin-clavulanate vs amoxicillin (treatment failure ~3% in both; RR 0.96 with CI crossing 1), while secondary infections are higher with amoxicillin-clavulanate; absolute event rates for serious harms are very low.
Implementability & Practicality 9.0
Both options are common, inexpensive oral antibiotics with straightforward dosing and minimal additional monitoring; switching to amoxicillin first-line is operationally easy.
Practice-Changing Potential 6.5
Provides strong real-world evidence suggesting amoxicillin is preferable to amoxicillin-clavulanate for uncomplicated cases due to similar failure and fewer secondary infections, but being observational may limit immediate guideline-level change for some clinicians.
5. 7.5
Calcium, vitamin D, or combined supplementation to prevent fractures and falls: systematic review and meta-analysis.
Overall: A large, high-quality meta-analysis of RCTs in mostly community-dwelling adults finds calcium and/or vitamin D provide little to no clinically meaningful reduction in fractures or falls, making it highly relevant and methodologically strong but with low demonstrated net benefit and a moderate-to-high potential to reduce routine supplement use in primary care.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 9.0
Addresses a very common primary-care question (routine calcium/vitamin D use for fracture/fall prevention) in largely community-dwelling adults.
Validity, Bias Control & Precision 9.0
Large systematic review/meta-analysis of randomized trials (69 trials; 153,902 participants) with independent duplicate processes, RoB 2 assessment, random-effects models, and GRADE; estimates are generally precise with moderate-to-high certainty.
Patient-Oriented Outcomes 9.0
Primary and secondary outcomes are clinically important (any fracture, hip fracture, vertebral/non-vertebral fractures, falls and number of falls).
Magnitude of Net Benefit 2.0
Effect sizes suggest little to no benefit: calcium RR 0.91 (CI crosses 1), vitamin D RR 1.00, and combined RR 0.91 with a very small relative effect; the abstract concludes absolute benefits do not meet clinically meaningful thresholds.
Implementability & Practicality 9.0
Interventions are simple, widely available, and low workflow burden in routine outpatient care (and the findings support avoiding unnecessary supplementation).
Practice-Changing Potential 7.0
High-certainty evidence of minimal benefit can credibly discourage routine supplementation for average-risk adults, though evidence is limited for higher-risk/residential-care groups so changes may be more targeted than universal.
6. 7.4
Exposure to Systemic Antimicrobials During Pregnancy and Risk of Miscarriage: A Population-Based Registry Study.
Overall: A large, methodologically careful registry study on miscarriage risk after early-pregnancy antimicrobial exposure with patient-important outcomes and actionable implications, tempered by observational confounding and lack of absolute risk estimates.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 8.0
Addresses a common primary-care decision (antibiotic selection in early pregnancy) using a broad, nationwide pregnancy cohort.
Validity, Bias Control & Precision 7.5
Very large, population-based registry cohort with time-varying exposure handling, lag to reduce reverse causation, and bias analyses; however, results remain observational with residual confounding by indication explicitly suggested.
Patient-Oriented Outcomes 9.0
Miscarriage is a clearly patient-important outcome and is the main endpoint, with registry-based capture.
Magnitude of Net Benefit 5.5
Reports elevated relative risks (HR ~1.5–2.0) for several agents, but provides no absolute risk differences and notes that underlying infections may drive associations, limiting confidence in net clinical impact.
Implementability & Practicality 8.0
Findings can be applied directly to outpatient prescribing choices (e.g., reassurance for nitrofurantoin/amoxicillin, caution with others) without new tests or complex workflows.
Practice-Changing Potential 6.5
May shift antibiotic selection or counseling in early pregnancy, but the nonrandomized design and stated potential confounding make it less definitive as a standalone practice changer.
7. 7.4
Vonoprazan-Tetracycline Dual Regimen as Rescue Therapy for Helicobacter pylori Infection: Randomized Controlled Trial.
Overall: A randomized noninferiority trial in previously treated H. pylori showed similar eradication to bismuth quadruple therapy but far fewer adverse events, no discontinuations, and better adherence with vonoprazan–tetracycline dual therapy, supporting a meaningful practical advantage despite open-label design and patient-centered outcomes being limited mainly to tolerability.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 7.5
H. pylori eradication (including salvage after failure) is a common outpatient problem that often involves primary care diagnosis, treatment, and referral decisions, though salvage regimens may be gastroenterology-led in some settings.
Validity, Bias Control & Precision 7.5
Prospective randomized controlled noninferiority trial with a clear comparator and reported CIs; however it is open-label (potential reporting/assessment bias for adverse events and adherence) and the abstract does not describe allocation concealment or blinding of outcome assessment.
Patient-Oriented Outcomes 6.5
Eradication is clinically meaningful but is not a direct patient-centered endpoint (e.g., symptoms, ulcer complications, cancer outcomes). Adverse events, discontinuation, and adherence are patient-relevant and clearly reported.
Magnitude of Net Benefit 8.5
Eradication rates were noninferior (~90% in both groups) with substantially fewer treatment-emergent adverse events (10.9% vs 45.7%) and fewer discontinuations (0% vs 8.6%), indicating a clearly favorable benefit–harm and burden profile for the simplified regimen.
Implementability & Practicality 7.0
A 14-day dual regimen is simpler than quadruple therapy and showed higher adherence, but it still requires multiple daily dosing (tetracycline TID) and access to vonoprazan may vary; no information is given on cost or availability barriers.
Practice-Changing Potential 7.5
If vonoprazan is available, a simpler salvage regimen with similar eradication and markedly fewer adverse events could change rescue-therapy choices; however generalizability beyond the studied context and longer-term clinical outcomes are not addressed in the abstract.
8. 7.3
Dose-Response Associations of Intermittent Lifestyle Physical Activity Micropatterns and Incident Type 2 Diabetes.
Overall: A large prospective accelerometry study links brief activity “microbouts” to lower incident diabetes risk with practical implications, but observational design and lack of absolute effects limit certainty and immediate practice change.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 8.5
Addresses prevention of incident type 2 diabetes in non-exercising older adults, a common primary-care target; findings could inform counseling on feasible activity in daily routines.
Validity, Bias Control & Precision 6.0
Large prospective cohort with accelerometer-derived exposure and linked records for outcomes, but observational design leaves substantial residual confounding/selection bias risk despite multivariable adjustment; CIs are reasonably tight for key estimates.
Patient-Oriented Outcomes 8.5
Outcome is incident type 2 diabetes ascertained via health records, which is clinically meaningful and patient-important rather than a physiologic surrogate.
Magnitude of Net Benefit 6.5
Associations suggest sizable relative risk reductions (e.g., HR ~0.64), but absolute risk reduction is not reported and causality is uncertain; proposed activity pattern likely has low harm and modest burden.
Implementability & Practicality 8.0
Intermittent brief bouts of moderate-to-vigorous activity during routine life are generally low-cost and scalable, though translating accelerometer-defined “bouts” into simple patient instructions may require coaching.
Practice-Changing Potential 6.5
Could influence how clinicians counsel sedentary patients (microbouts as an alternative to structured exercise), but evidence is not randomized and may be insufficient alone to shift guidelines.
9. 7.3
Effectiveness of Nirmatrelvir/Ritonavir for Outpatients in the Era of Omicron, Vaccination, and Previous Infection: A Meta-analysis.
Overall: A very large observational meta-analysis suggests nirmatrelvir/ritonavir lowers hospitalization and mortality in outpatients, with clinically meaningful absolute benefit mainly in higher-risk groups, though residual confounding and unreported harms limit certainty about net benefit.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 9.0
Addresses outpatient COVID-19 treatment decisions (nirmatrelvir/ritonavir) that are commonly made in primary care, with attention to contemporary Omicron-era and previously vaccinated/infected populations.
Validity, Bias Control & Precision 7.0
Very large meta-analysis (47 adjusted cohort studies; >10 million patients) with effect estimates and CIs/prediction intervals, but all included studies are observational, so residual confounding and treatment-selection bias remain possible.
Patient-Oriented Outcomes 9.0
Uses clearly patient-important endpoints (hospitalization and mortality), including both all-cause and COVID-19–specific outcomes.
Magnitude of Net Benefit 6.0
Relative risk reductions are substantial, but absolute benefit varies markedly by baseline risk (e.g., estimated NNT ~1148 low risk vs 20 high risk); harms/burdens are not presented in the abstract, limiting net-benefit assessment.
Implementability & Practicality 7.0
Intervention is a standard outpatient antiviral regimen and therefore generally feasible to deploy in routine care, but the abstract does not describe practical barriers (e.g., monitoring, contraindications, access), so real-world friction cannot be fully judged.
Practice-Changing Potential 6.0
Provides updated Omicron-era effectiveness estimates and risk-stratified absolute benefits that can refine prescribing (especially for moderate/high-risk patients), but it is unlikely to radically change practice given existing uptake and guideline awareness.
10. 7.2
A hybrid type 1 trial of a digital behaviour change intervention for frailty prevention in community-dwelling older adults aged 60 years and older.
Overall: A small, single-community RCT of a digital behavior-change program in older adults shows improved frailty index and modest QoL gains with acceptable user feedback, but limited sample size and incomplete reporting of key trial conduct details constrain confidence and immediate practice-changing impact.
View 6 Criterion Scores
Primary-Care Relevance & Applicability 8.5
Targets community-dwelling adults ≥60 years and focuses on frailty prevention and health behaviors (activity, vaccination, medication optimization, diet, socialization) that align with common primary-care prevention goals.
Validity, Bias Control & Precision 6.5
Randomised controlled design with a comparator and 6-month follow-up supports validity, but the sample is small (n=60), single local council setting, and the abstract does not report blinding or attrition, limiting certainty and generalizability despite reported CIs.
Patient-Oriented Outcomes 7.0
Includes quality of life (EQ-5D-5L), which is patient-important, but the primary outcome is a frailty index (a composite risk/health-state measure) rather than a hard clinical endpoint (e.g., disability, falls, hospitalization).
Magnitude of Net Benefit 7.5
Shows directionally favorable and statistically supported changes in frailty index and a modest QoL improvement with CIs provided; however, absolute clinical meaning and harms/burdens (time, adverse effects, disengagement) are not quantified in the abstract.
Implementability & Practicality 6.5
A digital intervention with newsletters and optional face-to-face talks is plausibly deployable, but the abstract notes technical barriers and implies engagement/support needs that may complicate routine scaling, especially for older adults with variable digital access.
Practice-Changing Potential 7.5
Suggests a feasible prevention approach for an aging population with measurable improvements over 6 months, but evidence is preliminary due to small size and single-site recruitment, making it more suggestive than definitive for immediate widespread practice change.
Score Guide: 9-10 Exceptional 7-8 Strong 5-6 Moderate 3-4 Weak 1-2 Poor
Showing top 10 of 278

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