Skip to main content

Table 3 Sub-group analyses

From: The weekend effect: does hospital mortality differ by day of the week? A systematic review and meta-analysis

Subgroup Number of analyses OR (95% CI) p-value I2 (%)
All Analysesa 50 1.07 (1.03, 1.12) 0.002 97.6
Date of publicationb
 • Pre-October 13th 2015c 30 1.09 (1.02, 1.17) 0.009 97.5%
 • Post-October 13th 2015 20 1.06 (1.03, 1.09) < 0.001 84.0%
Study sample size
 • Less than 10,000 15 1.03 (0.89, 1.18) 0.72 57.9%
 • 10,000 to 100,000 15 1.08 (1.04, 1.13) < 0.001 57.1%
 • 100,000 to 1,000,000 8 1.08 (1.04, 1.30) 0.001 81.9%
 • Over 1,000,000 12 1.11 (1.02, 1.21) 0.015 99.4%
Database type
 • Administrative 41 1.09 (1.04, 1.15) < 0.001 98.0%
 • Clinical registry/audit 9 1.03 (0.98, 1.09) 0.25 57.8%
Measure of severity
 • No measure of severity 37 1.08 (1.03, 1.14) 0.004 98.2%
 • Measure of severity 13 1.06 (1.02, 1.10) < 0.001 43.5%
  1. a50 analyses from 34 published articles
  2. bDate of publication was used rather than submission as date of submission was not available for all articles
  3. cThe 13th October 2015 was the date of the Minister Of Health’s statement in the House of Commons about the weekend effect