Skip to main content

Table 2 Methodological assessment of included studies

From: Intestinal dysbiosis in preterm infants preceding necrotizing enterocolitis: a systematic review and meta-analysis

Assessment criteria

Millar 1996 [63]

De la Cochetiere 2004 [39]

Wang 2009 [11]

Mshvildadze 2010 [30]

Mai 2011 [31]

Smith 2012 [64]

Stewart 2012 [40]

Norman 2013 [34]

Torraza 2013 [56]

Morrow 2013 [33]

Zhou 2015 [37]

McMurtry 2015 [32]

Sim 2015 [35]

Warner 2016 [38]

Do inclusion/exclusion criteria vary across comparison groups?

Yesa

No

No

No

No

No

No

No

No

No

No

No

No

No

Is the selection of the comparison group inappropriate?

No

No

No

No

No

No

No

No

No

No

No

No

No

No

Were valid and reliable measures (outcomes) applied consistently across all study participants?

Yes

Yes

Yes

Nob

Yes

Yes

Noc

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Are any important primary outcomes missing from the results?

No

No

No

No

No

No

No

No

No

No

No

No

No

No

Are results believable given the study limitations? (overall quality of the study)

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

To assess confounding

Any attempt to balance allocation between groups (e.g., stratification, matching)

No

Yes

Yes

Yes

Yes

No

No

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Were important confounding variables (gestational age or birth weight) taken into account in the design and/or analysis (by matching, stratification, multivariate analysis, etc.)

No

Yes

Yes

Yes

Yes

No

No

Yes

Yes

Yes

Yes

Yes

Yes

Yes

  1. Assessing methodological quality by the key components of study design is recommended over assigning quality scores in observational studies. The inclusion and exclusion criteria were similar in cases and controls in all included studies except Millar et al., where one intestinal sample along with fecal samples was included in the NEC group. Molecular methods of microbiome analysis were reported for all patients in all studies except Mshvildadze [30], Stewart 2012, where data for only a subset of patients were reported. All studies had appropriate comparison groups, reported all of our pre-specified outcomes and results “were believable given the study limitations.” Ten out of 14 included studies addressed the issue of confounding by matching or stratification to avoid known confounders
  2. aOne NEC sample was tissue sample and was obtained post-mortem
  3. bDGGE was performed in 23 infants and pyrosequencing was performed in a subset of 12 infants (6 of whom had NEC or sepsis)
  4. cMolecular assessment of stools by DGGE was assessed in only 27 infants out of 38 (16 infants with NEC or sepsis and 11 control infants)