Skip to main content

Table 2 Methodological quality of included studies

From: Epidemiology, prevention, and treatment of new-onset atrial fibrillation in critically ill: a systematic review

Author, year [ref]

Main outcome

Study design

Risk of bias

Indirectness

Imprecision a

Overall quality

Bender JS, 1996 [49]

Etiology/outcome

Prospective (0.5), observational (2)

AF diagnosis unclear (−1)

Mortality not evaluated (−0.5)

 

Very low (1)

Seguin P, 2004 [50]

Etiology/outcome

Prospective (0.5), observational (2)

   

Low (2.5)

Seguin P, 2006 [51]

Etiology/outcome

Prospective (0.5), observational (2)

 

Only trauma patients (−0.5)

 

Low (2)

Seguin P, 2006 [51]

Treatment

Prospective (0.5), observational (2)

 

Absence of definition for conversion time (−1)

N = 16 (−1)

Very low (0.5)

Arora S, 2007 [52]

Etiology/outcome

Prospective (0.5), observational (2)

  

N = 61 (−1)

Very low (1.5)

Christian SA, 2008 [26]

Etiology/outcome

Observational (2)

Retrospective diagnosis of AF (−1)

Only septic patients (−0.5)

 

Very low (0.5)

Meierhenrich R, 2010 [53]

Etiology/outcome

Prospective (0.5), observational (2)

   

Low (2.5)

Meierhenrich R, 2010 [53]

Treatment

Prospective (0.5), observational (2)

 

Unknown efficacy of each treatment (−1)

N = 49 (−1)

Very low (0.5)

Walkey AJ, 2011 [54]

Etiology/outcome

Observational (2)

Retrospective diagnosis of AF (−1)

Only septic patients (−0.5)

N = 49,082 (+2)

Low (2.5)

Kanji S, 2012 [55]

Etiology/outcome

Observational (2)

Retrospective diagnosis of AF (−1)

 

N = 2,895 (+1)

Low (3)

Kanji S, 2012 [55]

Treatment

Observational (2)

Retrospective diagnosis of AF (−1)

Unknown efficacy of each treatment (−1)

 

Very low (0)

Tongyaoo S, 2013 [56]

Etiology/outcome

Prospective (0.5), observational (2)

   

Low (2.5)

Makrygiannis SS, 2014 [57]

Etiology/outcome

Prospective (0.5), observational (2)

 

Mortality not evaluated (−0.5)

 

Low (2)

Balser JR, 1998 [60]

Treatment

Randomized (4)

Open label (−0.5), Allocation concealment unclear (−0.5)

Mortality not evaluated (−0.5)

N = 64 (−1)

Very low (1.5)

Sleeswijk ME, 2008 [61]

Treatment

Prospective (0.5), observational (2)

 

Mortality not evaluated (−0.5)

N = 29 (−1)

Very low (1)

  1. aSample size: N > 10,000 (+2), N ≥ 1,000 (+1), N < 100 (−1).