Internal Medicine

Many indirect comparisons with real-world controls omit causal frameworks and balance reporting.

This systematic review examined 55 single-arm drug trials that used real-world controls between 1987 and 2024. Only 29% of studies provided a rationale for using an indirect comparison to real-world data. Balancing-based methods were more recent and more likely to report post-adjustment balance, but reporting often relied on p-values. The authors caution that non-overlapping enrollment periods and limited balance reporting threaten the positivity assumption and validity of unanchored comparisons. 1

Protocol: individualized intraoperative systolic BP targets for older hip fracture patients.

This single-center RCT will randomize 180 patients aged 65–85 undergoing hip fracture surgery to individualized SBP targets (±10% of baseline) versus standard reactive care. All patients will have a universal MAP target of ≥65 mmHg maintained intraoperatively and in recovery. The primary outcome is postoperative organ dysfunction within 7 days; secondary outcomes include ICU/hospital stay and 30-day mortality. This is a protocol; results are needed before changing perioperative BP practice. 2

EHR-derived resident-sensitive quality measures show usable signals but attribution limits remain.

Using EHR data from five Canadian hospitals, investigators linked 132,291 admissions to 793 internal medicine residents to develop RSQMs. Low-value care transfusions occurred at a median of 0%, showing little variation across residents. Discretionary measures (second-line antibiotics, advanced imaging) showed wide resident-level variation. The authors note attribution and statistical reliability concerns that may limit inpatient RSQM use for high-stakes assessment. 3