30-Second Takeaway
- EHR-based behavioral nudges can meaningfully increase deprescribing of high-risk medications in older adults.
- Treat-to-target urate lowering in gout is associated with fewer major cardiovascular events, especially at lower urate thresholds.
- Current inpatient CGM devices are insufficiently accurate in the hypoglycemic range to guide insulin therapy without confirmatory testing.
- Ward team structure influences diagnostic error risk in deteriorating inpatients, with resident teams performing better than direct-attending models.
- Cumulative geriatric syndromes at admission strongly stratify 90-day mortality risk and should trigger targeted inpatient interventions.
Week ending January 31, 2026
Practical updates in inpatient safety, chronic disease targets, and care models for older adults
EHR nudges increase deprescribing of high-risk meds in older adults
This 3-arm randomized trial included 201 PCPs and 1146 patients aged 65 years or older on benzodiazepines, Z-drugs, or multiple anticholinergics. EHR-embedded precommitment messages increased deprescribing to 36.8% vs 26.8% with usual care (RR, 1.40; 95% CI, 1.14-1.73). Boostering reminders increased deprescribing to 34.3% vs 26.8% with usual care (RR, 1.26; 95% CI, 1.01-1.57). Deprescribing was defined as physician-directed discontinuation or tapering documented in the EHR during a median follow-up of about 9 months. Findings support integrating visit-triggered behavioral nudges to prompt deprescribing conversations for high-risk medications in older adults.
Treat-to-target urate lowering linked to fewer cardiovascular events in gout
This new-user cohort of 109,504 adults with gout starting ULT emulated treat-to-target vs non–treat-to-target urate strategies. Achieving serum urate <6 mg/dL within 12 months was associated with higher 5-year event-free survival and fewer major cardiovascular events (HR, 0.91; 95% CI, 0.89-0.92). Risk reduction was greater in people at high or very high cardiovascular risk and in those achieving <5 mg/dL (HR, 0.77; 95% CI, 0.72-0.81). Treat-to-target was also associated with fewer gout flares, while negative control outcomes were unchanged, supporting specificity of the findings.
Inpatient CGM is unreliable for hypoglycemia decisions on general wards
This systematic review included 9 studies of adults with diabetes on non-critical care wards, providing 465 paired hypoglycemic CGM and reference values. Mean absolute relative differences for CGM readings <70 mg/dL ranged from 7.6% to 53.3%, usually exceeding 15%. Pairing methods varied, but 8 of 9 studies reported mean absolute relative differences above 15% in the hypoglycemic range. The authors conclude current CGMs are too inaccurate in this range to guide in-hospital diabetes therapy without confirmatory capillary testing.
Resident teaching teams show lower diagnostic error risk than direct attending care
This multicenter retrospective study examined 1544 general medicine patients who transferred to ICU or died across 29 US hospitals. Patients were managed on resident teaching teams (63%), direct attending teams (29%), or APP-led services (9%). Direct attending care was associated with a higher risk of any diagnostic error vs teaching teams (mRR, 1.36; 95% CI, 1.04-1.68). Diagnostic errors, adjudicated by two-physician chart review, included missed opportunities regardless of harm and a subset causing temporary or permanent harm or death.
References
Numbered in order of appearance. Click any reference to view details.
Additional Reads
Optional additional studies from this edition.