Table 5.
Assessment of risk of bias in studies synthesized in the systematic review on long-term non-invasive ventilation in infants using the Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I) tool (18).
| First author, year | Confounding | Selection | Measurement of intervention | Missing data | Measurement of outcomes | Selection of reported results | Overall risk of bias (RoB) assessmenta |
|---|---|---|---|---|---|---|---|
| Obstructive sleep apnea | |||||||
| Downey (20), 2000 | Moderate | Moderate | Serious | Serious | Serious | Serious | Serious |
| Guilleminault (21), 1995 | Serious | Serious | Serious | Serious | Serious | Moderate | Serious |
| Harrington (22), 2003 | Moderate | Moderate | Serious | Moderate | Moderate | Moderate | Serious |
| Leonardis (23), 2013 | Moderate | Serious | Moderate | Serious | Serious | Moderate | Serious |
| Liu (24), 2012 | Serious | Serious | Moderate | Moderate | Moderate | Serious | Serious |
| McNamara (27), 1995 | Moderate | Moderate | Moderate | Serious | Moderate | Moderate | Serious |
| McNamara (28), 1999a | Moderate | Moderate | Moderate | Moderate | Moderate | Moderate | Moderate |
| McNamara (29), 1999b | Moderate | Moderate | Moderate | Moderate | Moderate | Moderate | Moderate |
| Robison (10), 2013 | Moderate | Moderate | Serious | Serious | Serious | Moderate | Serious |
| Rosen (30), 2010 | Moderate | Serious | Serious | Serious | Serious | Serious | Serious |
| Pierre Robin sequence | |||||||
| Amaddeo (31), 2016 | Serious | Serious | Serious | Moderate | Serious | Moderate | Serious |
| Kam (35), 2015 | Moderate | Moderate | Serious | Serious | Moderate | Moderate | Serious |
| Leboulanger (36), 2010 | Moderate | Moderate | Serious | Moderate | Moderate | Moderate | Serious |
| Goudy (34), 2017 | Serious | Serious | Serious | Moderate | Serious | Moderate | Serious |
| Laryngo-tracheomalacia | |||||||
| Essouri (38), 2005 | Moderate | Moderate | Moderate | Moderate | Low | Low | Moderate |
| Fauroux (39), 2001 | Moderate | Moderate | Moderate | Serious | Moderate | Moderate | Serious |
| Shatz (40), 2004 | Moderate | Serious | Serious | Serious | Serious | Moderate | Serious |
| Zwacka (41), 1997 | Serious | Serious | Serious | Serious | Serious | Serious | Serious |
| Spinal muscular atrophy type 1 | |||||||
| Bach (43), 2000 | Serious | Serious | Serious | Serious | Serious | Serious | Serious |
| Bach (44), 2002 | Serious | Serious | Serious | Serious | Serious | Serious | Serious |
| Bach (46), 2007 | Serious | Serious | Serious | Serious | Low | Moderate | Serious |
| Birnkrant (48), 1998 | Serious | Serious | Serious | Moderate | Serious | Serious | Serious |
| Chatwin (49), 2011 | Serious | Serious | Serious | Moderate | Moderate | Serious | Serious |
| Gregoretti (51), 2013 | Moderate | Moderate | Moderate | Moderate | Moderate | Moderate | Moderate |
| Lemoine (53), 2012 | Moderate | Serious | Serious | Moderate | Moderate | Moderate | Serious |
| Ottonello (54), 2011 | Moderate | Serious | Serious | Moderate | Moderate | Moderate | Serious |
| Vasconcelos (56), 2005 | Serious | Serious | Serious | Serious | Serious | Serious | Serious |
| Congenital hypoventilation syndrome | |||||||
| Hartmann (61), 1994 | Serious | Serious | Serious | Serious | Serious | Serious | Serious |
| Noyes (63), 1999 | Serious | Serious | Serious | Serious | Serious | Serious | Serious |
| Ramesh (64), 2008 | Moderate | Serious | Serious | Moderate | Serious | Serious | Serious |
| Tibballs (65), 2003 | Moderate | Serious | Serious | Moderate | Serious | Serious | Serious |
| García Teresa (60), 2017 | Serious | Serious | Serious | Moderate | Serious | Serious | Serious |
| Khayat (62), 2017 | Serious | Serious | Serious | Moderate | Serious | Moderate | Serious |
Low risk of bias—study is comparable to a well performed randomized trial within that domain. Moderate risk of bias—study is sound for a non-randomized study, but is not considered comparable to a well performed randomized trial within that domain. Serious risk of bias—study has some important problems within that domain. Critical risk of bias—the study is too problematic in this domain to provide any useful evidence on the effects of intervention. No information—no information on which to base a judgment about risk of bias within that domain.
aCriteria set out by the ROBINS-I tool.